Using NetMiner

２
Copyright and Trademarks

Using NetMiner
Version 4.4

Copyright 2000-2018 by Cyram Inc.

Companies, names and data used in examples herein are fictitious unless otherwise noted.

All Rights Reserved. No part of this publication may be reproduced, stored in a retrieval system,

or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording,

or otherwise, without the prior written permission of Cyram Inc.

The information in this publication is provided for information only and is subject to change

without notice. Cyram Inc. assume no responsibility or liability for any loss or damage that may

arise from the use of any information in this publication. The software described in this book is

furnished under license and may only be used or copied in accordance with the terms of that

license.

NetMiner is a registered trademark of Cyram Inc. CYRAM and CYRAM logo are registered

trademarks of Cyram Inc.

Cyram Inc.

#904, U-Space 2B, 670, Daewangpangyo-ro, Bundang-gu, Seongnam-si, Gyeonggi-do, 13494,

South Korea

Tel: +82-31-739-8352

Fax: +82-31-739-8354

Electronic access: http://www.netminer.com/

How to reference NetMiner 4

３
Using NetMiner

In order to reference NetMiner 4, please employ the following bold-faced reference, which is

also found at NetMiner Menu: “Help >> About NetMiner 4”.

Cyram (2018). NetMiner 4.4. Seoul: Cyram Inc.

Technical Support

The NetMiner Help file will guide you to use NetMiner 4. Simply press F1 function key or use

NetMiner menu bar for triggering Help feature.

Cyram is committed to provide a reliable, high-quality, yet easy-to-use product. If you have any

problem of installing or using NetMiner, please contact us by mail, phone and fax or preferably

by email and World Wide Web.

Cyram Inc.

#904, U-Space 2B, 670, Daewangpangyo-ro, Bundang-gu, Seongnam-si, Gyeonggi-do, 13494,

South Korea

Tel: +82-31-739-8352

Fax: +82-31-739-8354

Email: netminer@cyram.com Web: http://www.netminer.com/

When reporting a problem on NetMiner, please include the following information:

 your name and e-mail address

 version number of NetMiner program

 your hardware and software configuration. (ex. O/S, JRE version, RAM, CPU, etc.)

 a description of steps to reproduce the problem

４
Contents
I. Introduction

II. Concepts

1. Feature and Key concept

2. Data Structure
3. Process
4. Work Environment

III. Tasks

1. Data Handling
2. Process Handling
3. Visual Exploration

Appendix.

５
Using NetMiner

６
 III.Tasks

I. Introduction
1. What is NetMiner 4

2. System Requirements

3. Installation and Software Update

4. About this Document

5. Sample Data Description

7
Using NetMiner

What is NetMiner
Cyram NetMiner is innovative software for Exploratory Analysis and Visualization of Network Data. It

provides the integration of standard social network analysis (SNA) methodologies with modern network

visualization (or graph drawing) techniques for exploratory data analysis (EDA).

Cyram NetMiner allows an interactive and visual exploration of your network data and facilitates the

detection of underlying patterns and structures of the network.

Cyram NetMiner is capable of analyzing large-sized networks having up to million nodes. It does not only

include 110 network analysis and visualization modules, but also provide general statistical analysis

modules, machine learning-based analysis modules and chart modules.

Cyram NetMiner is a GUI-and-script-based software package that provides both a GUI mode and a script

mode. A user can easily perform his or her desired tasks in the GUI mode without having to handle the

tasks using the script mode since the results and processes would be the same. If the user decides to handle

process more finely and precisely, he or she can use the script mode instead. Thus, NetMiner is easy to use

and allows further customizations and expansions.

Cyram NetMiner can be utilized for general research, teaching and professional social network analysis

purposes. Furthermore, it can be effectively used in various business fields where network-structural factors

critically influence performance such as:

 Intra- and inter-organizational networks

 Financial and Web networks

 Criminal or intelligence networks

 Telecommunication networks

 Transportation and Distribution networks

Cyram NetMiner provides data mining algorithms such as Naive Bayes and CART. These data mining

algorithms can be combined with social network analysis to make analysis more powerful.

8
 III.Tasks

Main features of NetMiner include:

- Expressive and Systematic Network Data Model

- Large Network Data Analysis

- Exploratory Network Analysis

- What-if Network Analysis

- Comprehensive Network Measures and Models

- Workflow Management for productivity

- Powerful Data Management

- Interactive Visual Analytics

- User-friendliness

- Built-in functions/modules for Statistical Analyses, Charts and Data Mining.

- Ready for Presentation ; Chart and Video Recording

System Requirements

- Operation System

- 32bit version (Tiny, Small, Medium, Large) : Windows 2000, Windows XP

- 64bit version (Huge): Windows Server 2000 x64, Windows Server 2003 x64, Windows

Server 2008 x64, Windows XP x64, Windows Vista x64, Windows 7 x64
- Memory :

- 32bit version (Tiny, Small, Medium, Large): 512 MB RAM Minimum

- 64bit version (Huge): depends on the data size and algorithm complexity

- Hard Disk

- 32bit version : 300 MB of available hard-disk space is required

- 64bit version (Huge): depends on the data size and algorithm complexity

- Display : Super VGA (1024x768) or higher-resolution monitor

- USB port : for plugging-in USB flashdrive-looking hardware license key.

- Optionally a Windows-supported printer may be needed for printing outputs.

9
Using NetMiner

Installation and Software Update

Installation

NetMiner 4 installation Process is simple. You can simply install and use NetMiner 4.

Full installation guide is included in the installation package so that you can refer to this guide

during the installation process. You can also find the full installation guide in the “Getting

Started with NetMiner”.

Software Update: using “Help >> Check Update” menu

• In order to use the feature of “Help >> Check Update” menu, your computer, in which

NetMiner 4 has been installed, should have an internet access.

• By clicking “Help >> Check Update” option in the main menu, NetMiner 4 will be

automatically updated with our latest version.

• If current version is latest, following message box will appear.

10
 III.Tasks

• Step 1: If new update is available, a message box will pop up in order to ask you to update

the software. Click “Yes.”

11
Using NetMiner

• Step 2: From the NetMiner server, update file will be downloaded automatically.

• Step 3: After the download, upon “Update Setup Wizard,” please click “Next.”

• Step 4: Review License Agreement before installing. If you agree with it, click “I Agree”
button.

12
 III.Tasks

• Step 5: You can see the installation progress. For the progress details, please click Show

details button.

• Step 6: NetMiner 4 is now updated to its latest version. If you check “Run NetMiner 4,”
NetMiner 4 will be launched on clicking “Finish” button.

13
Using NetMiner

About This Document

This document describes how to use NetMiner 4 for network analysis, exploration and

visualization. Here you will find all the explanation which you should know to employ full

features of NetMiner 4. This document does not include tutorial for the methodology of

network analysis itself, but it is meant to give you a guide for the software: NetMiner. This

document is organized in a way that network analysts can easily follow in a step-by-step

manner in order to effectively use NetMiner 4.

This manual is associated with specifically NetMiner version 4.1. Please visit NetMiner Web

Site periodically for an update,

http://www.netminer.com

▶ See also flash movie about topic of this manual on the website

Sample Data Description

Available Sample Data in NetMiner are:

1. Five Cyram DataSets

2. Five Wasserman and Faust DataSets

3. Ten Mark Newman DataSets

4. One MovieLens DataSet

We used Wasserman and Faust DataSets in http://www.insna.org/INSNA/data_inf.html ,Mark

Newman DataSets in http://www-personal.umich.edu/~mejn/netdata/ , MovieLens DataSet in

http://www.grouplens.org/.

MovieLens DataSet is dataset for the test of Data Mining modules. so, they have somewhat

different data structure or size from general network datasets.

14
 III.Tasks

You can find the sample data in the folder where NetMiner 4 is installed. The default

installation folder is as follows: C:\Users\(Username)\NetMiner\

Sample data folder is composed of sub folders as follows:

Cyram DataSets
01. 01.Org_Net_Tiny1
02. 02.Org_Net_Tiny2
03. 03.Org_Net_Tiny3
04. 04.Org_Net_Small
05. 05.Internet_Medium

Wasserman and Faust DataSets

01. Krackhardt’s High-tech managers
02. Padgett’s Florentine Families
03. Freeman’s EIES network
04. Countries trade data
05. Galaskiewicz’s CEOs and clubs

Mark Newman DataSets

adjnoun
as-22july06
celegansneural
dolphins
football
karate
lesmis
polblogs
polbooks
power

Cyram DataSets

15
Using NetMiner

01. Org_Net_Tiny1

Main Nodeset: Employee

Twenty-two employees in an organization (node size = 22). The Label of Main Nodeset is the
name of each employee. Main Nodeset is described by six attribute data: Education, Gender,
Job-Ranking, Department, Duration and Age.

1-mode Network Data

Five 1-mode Network data represent the relations among the nodes in the Main Nodeset: Work
Interact, Personal Friend, Personal Help, Personal Knows and Work Help.

Sub Nodeset: Clubs

Three clubs with which the twenty two employees are affiliated.
Each club is described by five attribute data: (Club) Name, Manager, Meeting Day, Meeting
Time and the number of members.
2-mode Network data: Club Affiliation
One 2-mode Network data describes which clubs the twenty two employees are affiliated with.

Sub Nodeset: Interest Items

Five interest items the twenty two employees are interested in.
Each interest item is described by one attribute data: category (of each interest item).
2-mode Network data: Interested In
One 2-mode Network data represents which interest item the twenty two employees are
interested in.

Sub Nodeset: Commodities

Ten commodities, which the twenty two employees purchased.
Each commodity is described by one attribute data: Mart (where the employees purchased the
commodity)
2-mode Network data: Purchase
One 2-mode Network data represents which commodities the twenty two employees purchased.

02. Org_Net_Tiny2

Main Nodeset: Employees

Thirty nine employees in an organization (node size = 39). The Label of Main Nodeset
is the ID of each employee. Main Nodeset is described by seven attribute data: Team, Gender,
Education, Position, Duration, Name and Age.

16
 III.Tasks

1-mode Network Data

Five 1-mode Network data represent the relations among the nodes in the Main Nodeset: Trust,
Knowledge, Influence, Cooperation and Friendship.

Sub Nodeset: Clubs

Seventeen clubs with which the thirty nine employees are affiliated.
Each club is described by two attribute data: (Number of) Building and Room of each club.
2-mode Network data: Club Affiliation
One 2-mode Network data represents the affiliation relationship of thirty nine employees with
clubs.

Sub Nodeset: Interests

Seventeen interest items the thirty nine employees enjoy.
2-mode Network data: with friends and with coworkers
Two 2-mode Network data represent which interest item the thirty nine employees enjoy:
(enjoy) with friends and (enjoy) with coworkers.

Sub Nodeset: Movie Titles

Ten movie titles to which the thirty nine employees give score.
Each movie title is described by four attribute data: Director, User Rating, Category and
Company.
2-mode Network data: Score
One 2-mode Network data represents how much score the thirty nine employees give to each
movie title.

03. Org_Net_Tiny3

Main Nodeset: Employees

One hundred employees in an organization (node size = 100). The Label of Main Nodeset
is the ID of each employee. Main Nodeset is described by nine attribute data: Team, Gender,
Performance Level, Position, Job Satisfaction, Orgarnization Satisfaction and Department.

1-mode Network Data:

Five 1-mode Network data represent the relations among the nodes in the Main Nodeset:
Advice_From, Influence_From, Report_To, Talk About Innovative Ides With and Talk About
Personal Issue With.

17
Using NetMiner

Sub Nodeset: Issues

Four issues the one hundred employees are interested in.
Each issue is described by four attribute data: Importance (of each issue), Subject, Proposed
Date and Proposer.
2-mode Network data: Interested In
One 2-mode Network data represents which issues the employees are interested in.

04. Org_Net_Small

Main Nodeset: Employee

One thousand employees in an organization (node size = 1000). The Label of Main Nodeset is
the ID of each employee. Main Nodeset is described by twelve attribute data: Team, Sex, Join
as, Education, Performance Level, Position, Duration, Role, Name, Department, Age and
Group.

1-mode Network Data

Five 1-mode Network data represent the relations among the nodes in the Main Nodeset:
Cooperation, Influence, Knowledge, Personal and Trust

Sub Nodeset: External Ages

Fifty external agents, which the one thousand employees are contacted by.
Each agent is described by four attribute data: Sex, Company, Name and Age.
2-mode Network data: Contact
One 2-mode Network data represents by which agents the employees are contacted.

Sub Nodeset: Commodities

Twenty five commodities, which the one thousand employees purchased

2-mode Network data: Purchased on weekdays and Purchased on weekends

Two 2-mode Network data represent which commodities the employees purchased on
weekdays and weekends.

05. Internet_Medium

Main Nodeset: User

Ten thousand users (node size = 10000). The Label of Main Nodeset is the ID of each user.
Main Nodeset is described by eight attribute data: Frequency, Job, BlogOpenDateandTime, Sex,
Popularity, Education, Age and Location.

18
 III.Tasks

1-mode Network Data

Two 1-mode Network data represent the relations among the nodes in the Main Nodeset:
Blog_Buddy represents the buddy relation between the blogs of the user.
IM_Buddy: represents the buddy relation between the users who uses the instant messenger
service.

Sub Nodeset: Community

Five thousands on-line communities with which the users are affiliated
Each community is described by four attribute data: Open_date, Sub_category, Score and
Category.
2-mode Network data: Subscription
One 2-mode Network data represents which community the users subscribed.

Wasserman and Faust DataSets

These sample data are DataSets used in the text book “Social Network Analysis: Methods
and Applications" by Wasserman and Faust, 1994. Full descriptions of these data, including
references for the original sources of the data, can be found in Chapter 2 (pages 59- 66) and
Appendix B (pages 738-755) of Wasserman and Faust. There are five network examples:
Krackhardt's high-tech managers, Padgett's Florentine families, Freeman's EIES network,
Countries trade network, and Galaskiewicz's CEOs and clubs network. We prepared NetMiner
Project File for each example.

Mark Newman DataSets

We used some sample data in http://www-personal.umich.edu/~mejn/netdata. Full descriptions

about this data can be found in this site.

MovieLens DataSet

MovieLens data sets were collected by the GroupLens Research Project at the University of

Minnesota. This data set consist of 100,000 ratings(1-5) from 943 users on 1682 movies.

We extracted a 2-mode network data from this datasets. (@SOURCE:User ID, @TARGET:

Movie, @WEIGHT: Rating score(1-5))

19
Using NetMiner

II. Concepts
1. Feature and Key Concept

2. Data Structure

3. Process

4. Work Environment

20
 III.Tasks

Part 1. Feature and Key Concept

1. Expressive and Systematic Network Data Model

NetMiner has data model expressing various types of nodes, links, Node attributes, Link

attributes intergratively. So, NetMiner is able to express everything in the world.

2. Standardized Option and Output of Process

NetMiner arranges options of every process in several standardized categories. It also

standardize results of processes by classifying them into four types.

3. Session-based Execution of Process

NetMiner runs module by using session. Session contains selected module, input data, options

and the result of session.

4. Mix and Match of Numerical and Graphical Output

NetMiner provides results of analyses in graphical form and in numerical form at the same
time, which corresponds with each other appropriately.

21
Using NetMiner

5. Interactive Visual Exploration of Map

NetMiner allows additional action using results in graphical form without opening a new

session. Additional actions about graphic result divide into by clicking on map and by using

control panel by GUI standards. According to type of action, the actions divide into “pointing

action” and “shooting action”.

6. Integrating Output into Dataset

NetMiner provides easy and systemetical way to add results of analyses(in table form) to

existing data set or save them as a new data set.

7. Managing Data and Process History

NetMiner has a mechanism to track records of executed process and changed data. It helps the

user to find needed data and process easily, and to open that process and data and reuse them if

necessary.

8. GUI-and-Script-based Software with WYSIWYG Script.

NetMiner is GUI-and-Script-based Package software that provides both GUI mode and Script

mode. In other words, it has user-friendliness of GUI and expansibility of Script at the same

time.

22
 III.Tasks

Part 2. Data Structure

 Overview

 Main Features of NetMiner Data Structure

- Richness in representation of network data

Data structure of NetMiner is designed to represent almost every feature of network data.

You can import most of your network data into NetMiner.

- Integrated data structure

Each data item is integrated in a DataSet which is a basic unit of analysis and visualization

in NetMiner.

 Hierarchy of NetMiner Data Structure

A DataSet, composed of individual data items, is a basic unit for analysis and visualization in
NetMiner. A DataSet is used as an input data for all the analysis and visualization specific

modules in NetMiner. A DataSet is composed of four types of data items: Main Nodeset, Sub

Nodeset, 1-mode Network data and 2-mode Network data. A DataSet contains only one
Main Nodeset. But multiple 1-mode Network data can be contained in a DataSet. A DataSet
contains multiple Sub Nodesets and multiple 2-mode Network data for each Sub Nodeset can

be contained in a DataSet.

ProcessLogs which are generated in the analysis and visualization process can be managed

with the DataSet that is used for these analysis and visualization processes in a Workfile.

A Project contains independent multiple Workfiles. Number of Main Nodes in each Workfile
does not need to be the same.

23
Using NetMiner

In this way, the hierarchy of NetMiner data structure is as follow:

Project > Workfile > DataSet > Data items

Even if you perform a data analysis in NetMiner using only one data item, you need to

recognize the hierarchy of data structure in NetMiner.

The file extension ‘NMF’ represents Project which is in the highest level in the hierarchy of

data structure. If you ‘open’ a NMF file, you need to select a Workfile to use in the analysis and

then select data item(s) for each analysis and visualization specific module.

Each level in the hierarchy of data structure in NetMiner will be covered in the following topics.

24
 III.Tasks

1. Data Items

 About Network Data

A network is composed of nodes and relationships

among nodes.

Defining a group of nodes that compose a network as a

nodeset and defining the relationships among the nodes

- or a group of links - as a linkset, the nodeset and the

linkset on the left figure are as follows.

Nodeset = {A, B, C, D, E, F}

Linkset = {(A,B), (A,C), (B,D), (C,B), (D,F), (D,E)}

 Meta Matrix

The data structure of NetMiner can be explained through a Meta matrix, a matrix whose

rows and columns are nodesets. Given that there are nodesets A, B, C and D, the meta set can

be represented as shown below. Each cell of a meta matrix represents the relationship between

two nodesets. The matrix below shows all the relationships of each nodeset.

Let’s take a closer look by examining the row B in particular. Looking at the relationship
among the nodesets, it can be seen that it is related with all nodesets, including itself, centered

25
Using NetMiner

around B. In NetMiner, a row of a meta matrix is defined as a single workfile, and the nodeset

that acts as the reference is defined as a main nodeset, while other remaining nodesets are

defined as sub nodesets. In the figure below, B is the main nodeset, while A, C and D are sub

nodesets.

A network can be classified into either a 1-mode network or a 2-mode network depending on

the nodeset of which the network is composed. In a 1-mode network, the relationships are

within the same nodeset, while in a 2-mode network the relationships are between two different

nodesets. In the figure below, the cell at row B and column B is a 1-mode network while the

rest - the cells at row B and columns A, C and D - are 2-mode networks.

In NetMiner, there can be multiple networks composed of the same nodeset but with different

inter-node relationships. In NetMiner a group of such networks are called layers. Looking at it

with respect to the concept of meta matrices, layers can be considered to be a group of cells at
the same location with multiple meta matrices piled on top of one another.

26
 III.Tasks

 Main Nodeset and Sub Nodeset

Nodesets refer to a group of nodes that share some common attribute. In NetMiner, nodesets

are broken up into two types: main nodesets and sub nodesets. If there are multiple nodesets in

a single workfile, a single nodeset can be defined as the main nodeset; the rest will be defined

as sub nodesets. The main nodeset is a standard nodeset. Relationships within a nodeset can be

defined only for a main nodeset. For relationship between two nodesets, it is defined between a

main nodeset and a sub nodeset. This is to simplify user selection and processing by expressing

multiple nodesets in a single dataset and yet focusing on a single nodeset for analysis.

 Attribute of Nodeset

In NetMiner, multiple attributes can be defined for a node. Attributes can be number, text,

datetime, time, and even missing value. The four different types of attributes are represented

as shown below.

Node attributes are used for network data analysis or visualization. Specific modules of
NetMiner require attributes as input data when a network is analyzed.

Let’s take a look at how node attributes are used through an example. The following table

shows a nodeset composed of 6 people with 4 attributes, namely sex, age, department, and

position.

27
Using NetMiner

Using the attributes, only links that are above a specific value can be extracted to

reorganize the network. The attributes can also be considered to be groups, so that a network

of attributes can be built as well. The networks below show one that represents the

relationships between people whose age is over 30, and other that shows a reorganized

network by using a specific team for each node.

Relationships between people whose Relationship between teams

age is over 30

 1-mode Network and 2-mode Network

A network refers to a group of links of a specific nodeset or those of between nodesets.

A network is defined by using a specific nodeset as the reference, which is called the

reference nodeset for a particular network. A 1-mode network or a 2-mode network can be

explained according to the type of the reference nodeset.

In 1-mode network, the group is among nodes that are in the same nodeset, and only one

nodeset is involved. In NetMiner, networks that are based on a main nodeset are defined as 1-
mode networks.

28
 III.Tasks

2-mode networks are composed of a group of relationships among nodes that belong to two

different nodesets. They are based on two different nodesets. In NetMiner, networks that are

based on a main nodeset and a sub nodeset are defined as 2-mode networks.

In NetMiner data structure, one of the three nodeset must be selected as Main Nodeset. And

then the relation (1-mode Network) among the nodes in the Main Nodeset and the relations (2-

mode Networks) between the Main Nodeset and the other two nodesets (Sub Nodest) can be

represented as follows. The self-relations among the nodes in the two other nodesets and the

relations between those two other nodesets cannot be represented with NetMiner data structure.

29
Using NetMiner

 Representing a Network

A network can be represented as a Matrix Type or as a Edge List Type. In Matrix Type, all

nodes in a network can be represented including isolates. In Edge List Type, only the nodes

with links are represented. So the isolate nodes cannot be represented in Edge List Type. But

for this respect, Edge List Type is more efficient to represent a network. In Matrix Type, the

weight for each link is given by a value filled in matrix cell. In Edge List Type, a link is

represented as a row, each of which is composed of source node, target node and the weight.

1-mode Network 2-mode Network

Matrix

Type

30
 III.Tasks

Edge

List

Type

In NetMiner, all networks can be set in terms of direction, weight, and multiple link.

* Direction

* Weight

31
Using NetMiner

* Multiple link

Multiple Links in NetMiner are links (represented as multiple rows in Edge List Type)

which have identical source node and target node as follows.

In the screenshot above, you can see the identical source and target node pairs (CY269,

CY167). These are called Multiple Link in NetMiner. Multiple Links cannot be represented as a

Matrix Type. In Matrix Type, a weight for a link from source node to target node can be

represented only in a cell in the matrix. So, in NetMiner, a network data including Multiple

Link can be represented only as a Edge List Type. User can select whether or not to merge the

weights of the Multiple Links in the importing process using some methods as “Sum, Average,
Max and Min”. For example, if you select “Sum” as the merge option, the weight of the link

between CY269 and CY167 will be 7. And this network data can be represented in both

representation types. Even if you imported a network data including Multiple Links without

merge option, you need to merge the weights of the Multiple Links while running analysis and

visualization process.

 Attribute of Link

Multiple attributes can be defined for a link besides weight. The attribute of link is given to

each link not to the two nodes of the link. The following table shows an example of the

attribute of link.

32
 III.Tasks

In the table above, Time represents what time A meets B and Object represents for what

reason A meets B. These link attribute cannot be assigned to any node of each link. The

attribute of link must be assigned to each link itself. In this example, the link attribute as Time

and Object must be assigned to the “Meeting between A and B” not to the nodes “A or B”

Links can be set not only in terms of weight but multiple attributes, which include various types

such as number, text, datetime, time, and even missing value. Arbitrary link attribute values of

number type can be specified as weight.

 Representing Tree Structure

There are some occasions to handle tree structure of nodes while analyzing a network data. For

example, organizational chart of a firm shows a typical tree structure. When a user does

hierarchical clustering or community analysis, the tree structure of nodes is generated as a

result. It is usually represented as a dendrogram.

33
Using NetMiner

To represent a tree structure, NetMiner uses two kinds of Nodesets. One is the Main Nodeset

which is defined in your NetMiner dataset, and the other is ‘Tree Nodeset’ which is specially

designed for the tree structure. You can distinguish between the two Nodesets with an easy

example. Assume that there are two teams – Sales team and Marketing team in the

Sales&Marketing Department. Mary and Jesse belong to the Sales team and Willie and Bill

belong to the Marketing team. In this case, Mary, Jesse, Willie and Bill are the Main Nodes.

And ‘Sales team’, ‘Marketing team’ and the ‘Sales&Marketing Department’ are the Tree Nodes.

You can see that there is a hierarchy among the Tree Nodes. Sales&Marketing Department is

the ‘Parent Node’ of Sales team and Marketing team. In other words, Sales team and Marketing

team is the ‘Child Node’ of Sales&Marketing Department. In fact, we define the word ‘Tree

Structure’ as the ‘Parent – Child’ hierarchy in NetMiner 4. Tree Structure is represented with

the Sub Nodeset and the Sub Node attribute which shows the Parent Node of each Tree Node.

The whole hierarchy is divided in mainly two parts. One is the ‘Tree Structure’ which is

described in above paragraph, and the other is the ‘Inclusion Relationship’ which is the
relationship between Main Nodes and Tree Nodes. In NetMiner 4, Inclusion Relationship is

represented as a 2-mode Network, the relationship between Main Nodes and Sub Nodes. The

rules for making this specific 2-mode Network is described below.

1) One Main Node should be included in one Tree Node while a Tree Node can include

multiple Main Nodes.

2) A Main Node is usually included in the “leaf(terminal) node” – in the example, team level,

but sometimes a Main Node can be included in the “internal node” or even “root node” which

are not the “leaf node”. For example, a department manager will be directly included in the
Sales&Marketing Department.

34
 III.Tasks

In conclusion, NetMiner presents the hierarchy between Tree Nodes with Sub Nodeset and the

Sub Node attribute while the inclusion relationship is presented as a 2-mode Network. Thus, a

Sub Nodeset and a 2-mode Network can represent a whole Tree Structure and Inclusion

Relationship.

If the hierarchy is stored as some series of Main Node attributes, NetMiner 4 provides some

procedures to construct a Tree Dataset which includes ‘Tree Nodeset’, ‘Tree Structure’ and

‘Inclusion Relationship’. And there are some methods that utilize the Tree Dataset for analysis.
Representative menus are presented below:

Transform >> Mode >> Tree Construction

Analyze >> Position >> Expand/Collapse

35
Using NetMiner

2. DataSet

 Multiple Sub Nodeset

In a NetMiner dataset, there can be multiple nodesets.

Only a single nodeset can be defined as the main nodeset

and the rest are defined as sub nodesets. This is to

simplify user selection and processing by expressing

multiple nodesets in a single dataset and yet focusing on

a single nodeset for analysis.

In the figure to the left, the red box denotes the main

nodeset while the blue boxes denote sub nodesets.

 Layer

In NetMiner datasets, multiple networks that are based on the same reference nodeset but have

different relationships can be defined. For example, in a nodeset composed of six nodes {A, B,

C, D, E, F}, the following two 1-mode network data can be represented.

1-mode Network 1 (Fighting) = {(A,B), (A,C), (B,D), (C,E)}

1-mode Network 2 (Love) = {(A,C), (B,C), (D,E), (A,E)}

A group of networks that share a reference nodeset is called a network layer. There can be 1-

mode network layers where the main nodeset is used as the reference nodeset and 2-mode

network layers where the main nodeset and a specific sub nodeset are used as the reference

nodeset. If the reference nodeset is different, it can’t be defined as a layer. Therefore supposing
that there are multiple sub nodesets and 2-mode networks that are connected to each sub

36
 III.Tasks

nodeset, the number of layers that can be defined is the number of sub nodesets. That is, you

can’t group everything together and define as a layer just because it’s a 2-mode network.

In the figure to the left, the red box contains a group of

networks that represent employee relationships, and they

are 1-mode network layers. Although all 5 networks show

the relationships among employees, the definitions of the

relationships are different, such as ‘with whom does one

work together’ and ‘whose opinion has influence’. The

blue box contains a group of networks that represent what

leisurely activities the employees undertake, and are 2-

mode network layers. Although the two networks are the

same from the point that they represent the relationships

between employees and leisurely activities, they differ

depending on with whom the activities are undertaken.

Multiple network data in a nodeset can be integrated in the analysis and visualization process.

For example, you can extract the common links from the two 1-mode Networks and perform

analysis with only that links or merge the weights of each network for analysis.

37
Using NetMiner

3. Workfile

Workfiles are composed of a dataset and a process log. NetMiner uses the concept of a

workfile to manage a dataset as well as analysis process using the dataset simultaneously.

A process log shows traces of process execution, and is composed of Session, QuerySets, and

Selections. For Session, when a module is selected, a session is created and added to the list.

For QuerySets and Selections, when a query or a selection is saved, it is added to the list. Query

and Select are used when extacting only the parts in the current dataset that meet the criteria for

analysis.

For example, Degree Session contains the various

options and outputs of the Degree analysis module

using the current DataSet. QuerySet ‘Gender = male’ is

a condition for extracting nodes. If you are to analyze a

Module for male nodes, you can use this QuerySet

directly without extracting nodes in the analysis process

additionally. Selections are also can be used in the

similar way. Full descriptions are covered in “Tasks”.

38
 III.Tasks

4. Project

In NetMiner, a single data file (.nmf) corresponds to a single project, a 1-to-1 correspondence.

In a single project, there can be many workfiles.

There are two reasons for grouping multiple workfiles in a single project file. First is that in

NetMiner, there can be only one main nodeset in a workfile. In NetMiner, although a 1-mode

network that represents relationships among the nodes can be defined for a main nodeset, the

same is not possible for sub nodesets. In addition, when it comes to a main nodeset, although a

2-mode network that represents relationships with a sub nodeset can be defined, for a sub

nodeset, the same is not possible. For sub nodesets, only 2-mode networks that represent

relationships with a main nodeset can be defined. Because of this characteristic, for cases in

which a 1-mode network that represents relationships of nodes in a specific sub nodeset is

needed or when a 2-mode network that represents relationships between the sub nodeset and

another sub nodeset is needed, a new workfile is created in which the sub nodeset is made as

the main nodeset. As for the second reason, by managing multiple workfiles in a single project

file, management of similar data is made easier.

Workfiles are managed in a tree structure. A

new workfile is created when changes occur with

a main nodeset or when a data item to change is

linked with a previously executed process. To

reflect changes with the dataset in a new workfile,

by default it is set as a child of the current workfile.

The relationships between the parents and the

childs form the tree structure. The user can track

history of data changes using the tree structure

39
Using NetMiner

Part 3. Process

In NetMiner, a process refers to all actions that can be performed in NetMiner. There are

three types of processes: data handling process, analysis process, and visualization process.

The data handling process deals with creating and outputting datasets.

- Importing external text, Excel data or other types of network data files as NetMiner data.

- Creating results of analysis as new NetMiner data.

- Extracting parts of data using a query to create new data.

- Putting network data into a matrix and performing operations on it.

- Exporting NetMiner datasets as text or Excel files.

The analysis process deals with performing various network and statistical analyses with

NetMiner data. The visualization process refers to expressing network data using various

visualization algorithms.

The figure below shows the processes in NetMiner. The initials near the lines are the first

letter of the particular process. Specifically, ‘H’ refers to the data handling process; ‘A’ refers to

the analysis process; and ‘V’ refers to the visualization process.

The three processes are broken up into two broad categories depending on whether they are

performed in a standardized form. As the data handling process is performed in a non-

standardized way, it is classified as a non-session-based process, while the analysis process and
the visualization process are classified as session-based processes, as they are performed in a

40
 III.Tasks

standardized way. The following sections on modules and sessions will shed light on why a

standardized process is called a session-based process.

 Concept of Module

A standardized unit process is called a module. ‘Standardization’ refers to the fact that

control items for setting process options and types of process results are the same, for all

modules. In NetMiner, there are 6 control items and 4 result types, using which all modules of

NetMiner are expressed.

The 6 control items are: Filter, Input, Pre-process, Main-process, Post-process, and Output.

Input and Output are items that define the input data and the output for the particular process,

and they are mandatory for all modules. Filter is used to include only parts of input data (given

by specific conditions or selection) that meet specific criteria in the process. Although it is a
common item present in all modules, it is not mandatory. In contrast, Pre-process, Main process,

and Post-process are items that can be present or not present depending on the particular

module. They refer to the options related to how the process works.

The 4 types of results are the main report, table report, map report, and chart report. A main

report contains session information and brief abstract of the analysis results. A table report

contains numerical values in matrix format. A map report contains network visualization results.

Finally, a chart report contains graphical results in the form of ordinary charts.

For details on each module and interpretation on the results of each module, refer to “Module
Reference”.

41
Using NetMiner

 Concept of Session

A session is a user interface used when running a particular module, and includes selected input

data, processing options, actions performed, and the results. A module is not executed by itself

but through a session.

In the session window, on top there is a tab that shows the name of the selected module; on the

right there is a control panel where user options (input data, processing options, output) can be

selected; and in the center there is an output panel where the results are shown.

There are 3 session states: a module has been selected and a session has been open; options

have been set but execution is yet to be done; and the session has been executed.

The contents of a last-executed session are saved. Therefore if an existing session was opened,

options were set, and the session was closed without executing, when the session is opened

again the options set would be lost. If a module is selected, options are set, and the session is

closed without executing once, and then the session is opened again, the control panel wouldn’t

have any options set.

42
 III.Tasks

* Module name tab

In NetMiner, multiple sessions can be open simultaneously. But as they can’t be shown all at

once on the screen, tabs are used to switch between them. If a new session is opened when

there is already a session open, a new tab will be created to the right of the existing tab, and the

session indicated by the tab will be active. The tab of the currently active session is shown in

blue, and there is an ‘x’ for closing the active session.

If there is no room for any more tabs because too many sessions are open, an arrow icon with

the number of hidden tabs will be shown (see below), and when this is clicked, you can see a

list of the names of hidden sessions.

Closing a session doesn’t mean it gets deleted; you can open it again in the current workfile. So

it’s a good idea to close unnecessary sessions.

* Control panel
The control panel is a combination of 6 control items, according to the module. Generally

commonly-used control items are open in advance, while control items that are less frequently

used are closed. All items shown in the control panel can be used; closed control items can be

opened for use.

43
Using NetMiner

For example, the first figure on the

left is a control panel that appears

when the module ‘Analyze >

Centrality > Degree’ is selected. It is

made up of Filter, Input, Main Process,

and Output. Input and Main Process

are open while the other two items are

closed.

The second figure shows Output

open, which has been opened by

clicking.

The control panel doesn’t disappear when a session is executed but stays as it is with the

options set. Therefore the options set and the results can be seen at once, and after checking the

results the options can be immediately changed and executed again. When executing many

times with a single session open, the entire history doesn’t get saved but the results of the latest

execution and the options set are saved. Because of this, if you would like to save results of the
same module but with different options applied in a workfile, you have to open a new session

for execution.

* Output panel
The output panel shows the results of analysis or visualization process. If there are multiple sets

of results, the tabs on the bottom are used to switch between them. The tabs show the result

type and the result name. The result type is shown inside square brackets. [R] means main

report, [T] table report, [M] map report], and [C] chart report. There can be multiple instances
of the same result type in a single session.

44
 III.Tasks

In the main menu, the items that are related to a module are from Transform to Chart. The menu

and the module have 1-to-1 correspondence. Generally speaking, modules that belong to

Transform (making only links above a reference value as data or changing 2-mode data to 1-

mode data, etc.) can be considered data handling, but in NetMiner this process is considered as

analysis as well, so they are defined as modules.

- Transform: This category contains Modules for various data transformations. You can

extract nodes and links, change the values of link etc. using Modules in this category.

- Analyze: This category contains core Modules for analyzing network: Centrality, Cohesion,
Connection, Equivalence etc.

- Statistics: This category contains Modules for statistical procedures. Statistical measures for

network data are included. And you can get statistical results easily without moving to

external statistical packages.

- Visualize: This category contains Modules for network map visualization. Various layouts

for network map are available.

- Mining: This category contains Modules for data mining. (Based on machine learning

theories)
- Chart: This category contains Modules for visualizing charts.

45
Using NetMiner

The steps taken for module execution through a session are as follows.

Step1) Select a module.

Step2) Select the target data.

Step3) Set the options.

Step4) Execute the session.

Step5) Get the results.

 Difference between Analysis Session and Visualization Session

In the main menu, the modules in Visualize are related to the visualization process.

Visualization process modules are broken up into two: general modules and visualization

specific modules. Among sub-menus of Visualization menu, Layout and Drawing are general

modules while the rest are visualization specific modules. General modules and visualization

specific modules have differences in the execution process as well.

Step1) Select a module.

The session that appears when a general module is selected is different from the one that

appears when a visualization specific module is selected in terms of its make-up. When a

general module is selected, the session that appears will be as shown as below. There won’t be

anything in the output panel, and the process tab will be active in the control panel. The control
panel is made up of Process, Display, Select, Inspect and Video tabs.

46
 III.Tasks

When a visualization specific module is selected, you will see a session as shown below. In the

output panel, a map with nodes placed at arbitrary locations will be shown, and the control

panel will have the display tab active. The control panel is made up of Process, Display,

Select ,and Video tabs.

47
Using NetMiner

Step2) Select the target data.

In the general module, the target data for analysis is selected in Input in the process tab. But in

a visualization specific module, the target data for visualization is selected in Network of

Display.

Step3) Set the options.

When a general module is selected, the options for the particular module are shown in the

process tab, which you can set according to how you want the analysis to be done. In a

visualization specific module, the layout algorithm can be selected in the display tab, and you

can set the options for the algorithm.

Step4) Execute the session.

In an general module, click the ‘Run Process’ button in the process tab to execute the session.

In a visualization specific module, click the ‘Run Layout’ button in the display tab to execute

the session.

Step5) Get the results.

In an general module, a main report is outputted by default, and according to the particular

module, a table report, map report, and chart report are outputted in specific combination. In a
visualization specific module, only a map report is outputted.

48
 III.Tasks

 Non-Session Based Process

For non-session-based processes, the execution is done differently for each process. For most

cases excluding the query composer that extracts data through a query and the matrix calculator

that does matrix operations by putting network data in a matrix form, execution is done by the

way of a dialog box, which doesn’t leave any records of the options set because it gets closed

when the execution is done after setting the options. Therefore, in NetMiner, processes that are

not repetitive by nature, such as opening a file, importing external files, and exporting results to

external files are non-session-based, in contrast to analysis/visualization tasks.

The figure below shows a project file being open using the ‘File Open’ menu. As shown in the

49
Using NetMiner

figure on the left, when you click the OK button after selecting a file in the file selection dialog

box, the selected project will be open, as shown in the figure on the right.

50
 III.Tasks

Part 4. Work Environment

 Overview

• Main Window Segmentation

NetMiner User Interface is follows.

1. Title Bar

2. Main Menu

3. Tool Bar

4. Current Workfile

5. Workfile Tree

6. Output Panel

7. Control Panel
8. Status Bar

51
Using NetMiner

1. Title Bar

Title Bar represents the NetMiner Logo and the title of the Current Workfile.

2. Main Menu

Main Menu of NetMiner is composed of File, Edit, Data, Map, Tool, Transform, Analyze,
Statistics, Mining, Visualize, Chart, Window and Help. General and Visualization specific

modules are surrounded by the red box.

File Menu contains New, Open, Save, Import and Export etc. Edit Menu contains Copy, Paste,

Cut, Delete etc and Preference for setting global option. Data Menu contains sub menus for
handling data items. Map Menu contains various sub menus for handling network map and map

panel. Tools Menu contains Script Workbench, Query Composer, Matrix Calculator and Plug-in.

You can show and hide sub windows in Window Menu. General and Visualization specific
modules can be summarized as follows.

52
 III.Tasks

All the main menu items are not always activated. Activation of each main menu item depends

on the context of the current state of the program. For example, if ‘Workfile’ is not activated,

only ‘File’ , ‘Edit’ , ‘Tools’ menus are activated.

53
Using NetMiner

3. Tool Bar

Shortcut icons in Tool Bar can be categorized as follows, for easy access to frequently used

functions in NetMiner. Activation of each main menu item depends on the context of the

current state of the program. Thus, all the main menu items are not always activated.

New Project: Create a new project

Open Project: Open a project file(.nmf)

Save Project: Save the current project

Print: Print Main Report

Script Workbench

: Launch Script Workbench

Query Composer: Activate Query Session

Matrix Calculator

: Activate Matrix Calculator Session

Run Process: Run the analysis and

visualization session in process panel

Run Layout

: Draw again the network map using the

selected layout in display control panel.

Fit to screen

: Automatically resize the network map fit

to the size of window application

Zoom node(-)

: Decrease the size of nodes and the label

Zoom node(+)

: Increase the size of nodes and the label

54
 III.Tasks

Node and Link attribute Styling: Launch the dialogue box for node and link

Node Label: Select the node label

Link Label: Select the link label

Link Line: Show or hide the link

Link Width: Change the width of link

Link Head: Change the shape of arrow of the selected links

Directional Link Style: Select the directional link style.

Show Axis: Show or hide the axis of the 3D network map

Link Threshold: Show or hide links on the

network map using slide bar

Save Threshold to QuerySet: Save the

condition (weight of link) for showing

links on the network map as queryset.

Save Styles: Save the styles that are applied

to the current network map.

Save Coordinates: Save the coordinates of

the nodes on the network map as the

attribute data of main nodeset

Capture Map: Capture map to clipboard or

Save to image file. .bmp, .jpg, .png, .svg is

avaliable. .svg is an XML-based vector

image format and scaling the vector image

55
Using NetMiner

preserves the shapes.

Project View: Show or hide current

workfile and workfile tree

Help Contents: Launch Help Contents

56
 III.Tasks

4. Data Management Area

Data Management in NetMiner can be handled in Current Workfile or Workfile Tree area in

NetMiner. The red area in the following picture is for Current Workfile area, and the blue area

is for Workfile Tree. Current Workfile, in turn, represents the structure of Current DataSet and

the ProcessLog. Workfile Tree represents the data transformation history, and here you can

manage multiple Workfiles in this area.

 Current Workfile

Current Workfile includes checking and managing data items in Current DataSet (in the blue

box) and ProcessLog (in the red box). DataSet and ProcessLog can be selectively

displayed/hidden by selecting the check box at the top of the Current Workfile.

If you double click each data item in Current Workfile, you can see the content of the selected

data item in the right area (Data Editing Area). The Number of nodes (or links) and attributes
are represented at the data items. For example, [39*8] which is attached to the Main Node data

57
Using NetMiner

item ‘Employee’ represents that the Main Nodeset contains 39 nodes and 8 attribute data.

If you double click each item in Sessions, you can see the

options and outputs of the Sessions.

You can apply QuerySet and Selections using Filter Control

item in Process Control panel.

Full descriptions about data handling will be covered in

“Tasks – Data Handling”.

58
 III.Tasks

 Workfile Tree

You can see the structure of Workfiles in the Workfile Tree area as follows.

The Current Workfile is indicated as yellow icon. You can activate a selected Workfile as

Current Workfile by double clicking it.

59
Using NetMiner

5. Data Editing Area

Data Editing Area is where data editing work is done. If you open the selected data item in
Current Workfile, a Data Editing Session will be activated. Only one single Data Editing

Session can be activated for a data item by opening the data item in Current Workfile. Data

Editing area is composed of two parts: Data Editing panel and Data Control panel as follows.

 Data Editing Panel

You can perform various data editing work in the Editing Panel. The environment for data

editing in this panel has three View: Matrix View, Edge List View and Map View. According to

the type of data to edit, you can select one of them using the inner tab at the bottom of the

Editing Panel. Map View is available only for the 1-mode Network data.
In Matrix View, you can cut/copy/paste as you work in external spreadsheet program e.g. Excel.

60
 III.Tasks

In Edge List View, you can edit a network data that contains link attribute data or a network

data containing Multiple Link.

< Matrix View > < Edge List View> <Map View>

 Data Control Panel

Data Control Panel contains some Control items for editing work in Editing area. These are
Main, Value Label and Missing Value. Control items in Control panel are categorized according

to the data type to edit.

In Main Control item, you can check the network information,

select the number precision, choose link attribute to represents at

each cell in Matrix View and launch the Graph Editor for 1-mode

Network data. In Network Information, you can check whether

or not Multiple Links exist in the Current Workfile and whether

the Current Workfile is directed or not. You can represent number

in each cell to the sixth decimal points using Number Precision.

You can reorder the Main Nodes using Permutation Vector. You
can represent the value of link attribute in each cell in Matrix

View by selecting attribute of link in Choose Link Attribute. In

Matrix View the cells in the matrix can represent only one value

at one time, you need to select a value to represent. Default value

is the weight of each link, but you can change it. You can launch

61
Using NetMiner

Graph Editor by clicking the “Open Graph Editor” button. You can see a new inner tab named

“Map” at the bottom of the Editing Panel. Display Control Panel contains some simple Control

items like Zoom In/Out and Navigation.

You can assign a label to the original value using Value Label.
For example, you can associate each numerical value 1, 2, and 3

with user-defined labels,“CEO”, “Department Manager” and

“Team Manager," respectively. The analysis calculation

algorithm uses the original value for its internal computation, but the output result can be

represented using the label. In this way, it will be helpful to understand the output conveniently.

You can designate a selected value as missing value by using

Missing Value. The color of the cell representing the designated

value changes to red and the missing values are excluded from

the calculation.

62
 III.Tasks

6. Process Management Area

Process Management Area is where Process Sessions are activated. Main analysis and
visualization work is done in this area. You can create a Session by selecting main menu

representing the available analysis and visualization specific module: “Transform, Analyze,

Statistics, Mining, Visualize and Chart”. Multiple Process Sessions can be created for a Module.

You can also activate the created Sessions in Current Workfile.

Process Management Area is composed of two parts: Output Panel and Control Panel. Output

Panel is for browsing the generated outputs and the Control Panel is for control the outputs

using various Module-specific options.

 Output Panel

The generated outputs can be classified as four types: Main Report, Numerical Table,
Network Map and Chart.

63
Using NetMiner

Main Report is the default activated output generated for almost every general module. It
shows the summarized results of the analysis processing. The name of the Main Report is

common to all the general modules, “[R] Main”, at the inner tab named.

Numerical Tables contain various numerical results. Multiple Numerical Tables can be
generated for an general module. “[T]” is attached to the name of the Numerical Tables at the

inner tab.

For some general modules, Chart is generated as an output. You can see “[C]”, in front of the

name of chart output at the inner tab. Network Map is generated for most of the General
module in Analyze menu. Especially, Visualize Modules generate only network map. For

network map, you can see “[M]” in front of the name at the inner tab. And the name of the

inner tab for network map represents the layout algorithm for drawing the network map. If the

Network Map has a video object, 'Time slide bar' is on the output panel. For details on Video,

refer to “Task>>Recording Visual Exploration”.

You can browse the generated outputs in a Session by selecting the inner tab at the bottom of

the Output Panel as follows.

64
 III.Tasks

 Control Panel

Control Panel contains various Module-specific options for controlling the output results.

Control Panel is composed of five Panels: Process, Display, Select, Inspect, and Video.
Process Control Panel contains options that must be selected before running process and other

four Control Panels contain various options that can be used to control the generated network

map. These four Control Panels are activated only for the network map. You can check that the

font of the name of Control tabs is italic when there is not a network map on the Output Panel.

Each Control item can be collapsed or expanded as follows.

- Process Control Panel

Control items in Process Control Panel can be classified as follows.

1. Filtering

2. Standard Control items: Input, Pre-process, Main process, Post-process, Output

Filtering Control item contains user interface for selecting QuerySet or Selection to extract
nodes and links with specific condition.
Other five Control items (Input, Pre-process, Main process, Post-process and Output) are

65
Using NetMiner

standardized common process Control items. They represent the steps in data flow from input

to output for processing general module. You can select data item(s) from current DataSet as

input data in Input Control item. You need to select one or multiple data items according to the

Module. In Pre-process, Main process and Post-process, you need to select options for the

analysis algorithm. In Output Control item, you can select the output to generate and the
output format. For example, you can exclude some of outputs that require much time to

generate. Pre-process, Main process and Post-process Control items are not always required,

and when those are required, those Control items are shown in Process Control Panel.

- Display Control Panel

Display Control Panel contains various options for controlling the network map on the Output

Panel. Charts on the Output Panel are not controlled by the Control items in Display Control
Panel. These items control only the elements on the ‘network map’.

66
 III.Tasks

Display Control Panel is composed of five Control items: Navigation, Network, Layout, Node

Style and Link Style. Network Control item is contained in Display Control Panel only in

Visualize Modules. In Analyze Modules, you can select 1-mode Network data item in Input

Control item in Process Control Panel.

Navigation Control item is used to zoom in (out) or to navigate

the network map.

Network Control Item is used to select 1-mode Network data for

visualization. You can select and visualize multiple 1-mode

Network data at one time.

Layout Control item is used to

select layout algorithm and various

algorithm-specific options to draw

network map. You can run the

layout algorithm by hitting Run

Layout button.

67
Using NetMiner

Node Style Control item is used to select shape, color and size
of node and to select the node label. You can assign optimal

style to each node according to its attribute of node in the Node

Attribute Styling dialogue box that will be launched by hitting

the Node Attribute Styling button.

Link Style Control item is used to select width, color and arrow
shape of link and to select the link label. You can give optimal

style to each link according to the weight and attribute of link in

the Link Attribute Styling dialogue box that will be launched by

hitting the Link Attribute Styling button.

- Select Control Panel

You can select nodes with specific condition or show links on the network map according to the

weight of link in Select Control Panel. Three Control items are available: Condition, Selection

and Link Threshold.

In Condition Control item, you select nodes with specific condition. You can narrow down
nodes with various conditional restrictions on the attribute of Main Nodes.

68
 III.Tasks

You can check which nodes are selected, and you also can manually select individual node in

the node list. It is useful when there are too many nodes to select a specific node using mouse

on the network map. The selected nodes can be saved as Selection and the saved Selections can

be loaded and be applied to the current network map. You can also create a new Workfile with

the selected nodes.

Link Threshold Control item is used to show or hide links

with selected weight threshold. You can save the condition for

showing links on the network map as a QuerySet. You can

also save the nodes and links on the network map as a new

Workfile. Various options are available in selecting links on

the network map.

- Inspect Control Panel

Inspect Control Panel contains Module-specific Control items that are designed to inspect the
analysis results of that Module in the network map interactively.

For example, Shortest Path Module generates the Distance Matrix as a numerical table that

represents how many nodes must be passed from the source node to the target node using the

Shortest Path between these two nodes. Network map is generated at the same time with the

Distance Matrix for inspecting the Distance Matrix in the network map. If you select a source
node and target node to inspect, you can see the two nodes and the shortest path in the network

69
Using NetMiner

map. You can understand the output result intuitively using the Control items in Inspect Control

Panel interactively.

- Video Control Panel

You can record 2D or 3D visual explorations in Video Control Panel. Two control items are
available : Record and Edit.

Record control item is used to creation of a new video segment.

Record New : Start a new video recording

Record Insert : Start a video recording which is to be inserted at the current navigation pointer.

Record Stop : Stop recording

Edit control item is used to edit recorded video.

70
 III.Tasks

Undo / Redo : You can undo/redo the previous editing step.

Change Speed : You can change play speed of the selected range by adjusting whole play time

value or ratio.

Dissolve : Connect recording actions more smoothly based on the beginning and end images in

the selected range.

Delete : Delete selected range of video

7. Status Bar

Status Bar at the bottom of NetMiner Main Window represents the name and the status of the
current Module. The status of the Module can be one of the following four status: Ready,

Started, Finished and Cancelled.

Progress Bar as follows is activated when the Module is started. You can check how much
time it will take to finish the current Module in the Progress Bar, and you can decide to stop

running algorithm by hitting the red Stop button. This feature of the analysis environment
makes the NetMiner interruptible by users and is useful especially in the time of analysis with

large scale data using algorithms requiring high time-complexity.

71
Using NetMiner

8. Preference (Global Option)

You can give some basic parameters as global option for NetMiner in “Edit >> Preference”

menu as follow.

Details about each tab in Preference dialogue box as follows.

 Application

Application tab handles some options regarding the common application options and editor

options. If the Common >> Auto-Update option is checked, your NetMiner 4 is automatically

updated to the newest version when you start the program.

You can designate the default folder for Open, Save, Import, Export dialogue box in Common

>> Folder Path as follows. The latest folder which you used to handle the data is the default

option for the unchecked item – Save and Export option in the screenshot.

In the Editor >> Number Precision option, you can set which number of decimal places should

be presented in the editor. Maximum value is 6.

72
 III.Tasks

 Session

Session tab contains some options for handling session. You can designate the number of nodes

for the ‘map alert’. Map alert is a dialogue box for option to contain or exclude map output in a

generated outputs especially with many nodes as follows.

73
Using NetMiner

You can set option to add symmetric network to 1-mode network.

You can also give some basic parameters for the layout algorithm and initial coordinates for

visualization layout control panel as follows.

74
 III.Tasks

 Node

You can give default style for nodes on the network map in Node tab. Various item categories

are used in Inspect inner tab in Analyze module. Details for each item categories are covered in

Analyze section in NetMiner Module Reference.

 Link

You can give default style for links on the network map in Link tab. Various item categories are

used in Inspect inner tab in Analyze module. Details for each item categories are covered in

Analyze section in NetMiner Module Reference.

75
Using NetMiner

 Font

You can give default font style in Font tab as follow.

- Common – Normal: font style for node label on network map

- Chart – Matrix Diagram: font style for label in matrix diagram

 Background

You can give default background style in Background tab. Three options are available: single

color, gradient color and picture file as follows.

76
 III.Tasks

 Other

Other options are available. Only one item is available Chart – Matrix Diagram, which is for

selecting two colors for maximum and minimum value in matrix diagram as follows.

77
Using NetMiner

III. Tasks
Part I. Data Handling

Part II. Session Based Process Handling

Part III. Visual Exploration

78
 III.Tasks

III. Tasks
Part 1. Data Handling

1. Handling Project

2. Handling Workfile

3. Creating Data

4. Importing Data File

5. Importing DB Tables

6. Exporting Data

7. Editing Data

8. Transforming Data Using Matrix Calculator

9. Using Output Tables Recursively

10. Querying Data

12. Available Actions in “Current Workfile”

13. Available Actions in “Workfile Tree”

79
Using NetMiner

1. Handling Project

 Opening Project

You can check that Blank Project (“Untitled Project”) is created automatically, when you

launch NetMiner. You can start NetMiner by creating new Project or opening the saved Project.

In this topic “Project Open” is covered. Creating Project is covered in “Data Handling –

Handling Project – Creating Project.”

• Open vs. Import

Prior to this topic “Opening Project,” you need to know the difference between “Open” and

“Import” in NetMiner. “Open” in NetMiner is a process for opening the NetMiner Project file

with NMF extension only. “Import” is a process for importing external source files (TXT, CSV),

NetMiner Workfiles, other software files and DB Tables. In case of file importing, you can

import files with following file formats.

- Generally Used TXT file: CSV, TXT

- Microsoft Excel file: XLS

- Prior version of NetMiner file: NTF

- Compressed file of Workfiles of NetMiner: NCW

- UCINET file:DL

- Pajek file: NET, VEC

- StOCNET file: DAT, TXT

- GML file: GML

Full descriptions about importing data are covered in “Importing Data File” or “Importing DB

Tables”. Workfile importing is covered in “Handling Workfile – Importing Workfile”.

80
 III.Tasks

• Open Project

Select main menu “File >> Open (CTRL+O)” for launching dialogue box for selecting

NetMiner Project file. You can do this process easily just by selecting the icon in the Tool Bar

as follows.

In the file selection dialogue box, select the “02.Org_Net_Tiny2.nmf” file in the following

folder: “C:\Users\(Username)\NetMiner\SampleData\Cyram DataSet\02.Org_Net_Tiny2\”

You can see the selected project is opened as follows. Red lined box in the following picture

represents the data structure of the opened Project file.

81
Using NetMiner

You can open the Project file just by double clicking the Project file (.NMF) in the folder that

contains the Project file because NMF file is associated with “NetMiner.exe”. You can check

the recently opened Projects (maximum five Projects) in File >> Recent Projects menu. You

can open these Projects conveniently using this menu.

 Creating Project

There are two ways for creating new Project in NetMiner. One is creating Project as a “Blank

Project” and the other is as a “Singleton Project”. Blank Project does not contain Workfile and

is ‘blank’. You need to create Workfile, data items and data for each data item in the Blank

Project. Singleton Project contains a Workfile which contains a 1-mode Network data item with

1 node. Singleton Project makes it possible to use “Graph Editor” directly in the 1-mode

Network data editing session. So, if you want to create a 1-mode Network data with small

number of nodes in the graphic user interface using “Graph Editor”, Singleton Project is

recommended as a way for creating new Project. Blank Project is an option for a user who

wants to compose his own data structure from the ‘blank’ state.

Select main menu “File >> New >> Project” for launching dialogue box for selecting the type

for creating new Project as follows.

• Creating Blank Project

Select “Blank Project” in the dialogue box for creating new Project. You can see the created

82
 III.Tasks

‘blank’ Project as follows.

Now you can create Workfile and create data for each created data item and import data for

composing a DataSet. Full descriptions for each topic are covered in the following sections

about “Data Handling”.

• Creating Singleton Project

Select “Singleton Project” in the dialogue box for selecting the type for creating new Project.

You can see the “Untitled Workfile” in the Workfile Tree area and the “Untitled Nodeset” and

“Untitled Network” in the Current Workfile area which represents the “Main Nodes” and “1-

mode Network data” each. Double click the 1-mode Network data named “Untitled Network”,

you can see there is only 1 node without link in that data as follows. Now you can use “Graph
Editor” for creating nodes and links just by clicking the “Open Graph Editor” button at the

83
Using NetMiner

right side of the panel. Full description about Graph Editor is covered in “Editing Data –

Editing Data Using Graph Editor”.

 Saving Project

A Project must be manually saved by user. NetMiner does not save the project automatically

when you exit the program. If you exit the program without saving the Project, all the works

done in NetMiner will be lost.

You can save Project by selecting main menu “File >> Save (or Save as)”. The saved file is a

Project file with NMF file extension. You can launch NetMiner with the selected Project file

just by double clicking it.

You can see the following dialogue box for selecting saving option if you exit NetMiner.

84
 III.Tasks

* Auto Saving.
If the program is terminated unexpectedly in abnormal way, the current Project is saved

automatically as a temporary file. You can recover and open the auto-saved Project when the

program is launched again. You can see a dialogue box representing the auto-saved Project list

as shown below. Select a Project to open and click Open button.

 Closing Project

You can close current Project by selecting main menu “File >> Close”. This action does not

terminate the NetMiner 4, but literally close the Project. You can create or open other Projects.

You can see the following dialogue box for selecting saving option when closing current

Project.

85
Using NetMiner

86
 III.Tasks

2. Handling Workfile

 Creating Workfile

Workfile in NetMiner is a basic data unit for processing analysis and visualization. It is
composed of two parts: DataSet and Process (Sessions, QuerySets, and Selections). You need to

create Workfile to create data or import external source files for data analysis and visualization.

Select main menu “File >> New >> Workfile” for launching the dialogue box for creating

Workfile as follows.

You can see the created Workfile as follows. There is a Workfile in the Workfile Tree with violet

colored icon. In the Current Workfile area you can see the blank data item and process item.

Now you can create or import data in the created Workfile. Full description is covered in

following sections about creating and importing data.

87
Using NetMiner

 Importing Workfile

Prior to this topic, it is recommended that you be familiar with the concept of “Workfile” in

NetMiner at “Concepts – Data Structure”. Basically importing/Exporting Workfile(s) is similar

to Open/Save Project. You can select Workfile(s) to import or export in this process. In

importing process, you can select Workfiles to import from the Workfiles contained in a Project

with NMF or NCW file format. NCW file format is compressed Workfiles data, which was

used in the previous version of NetMiner 4. NetMiner 4 can import NCW files for backward

compatibility. You can export the selected Workfiles in the current Project as a NMF file. The

Tree Structure among the selected Workfiles in the Workfile Tree is maintained in the exported

file.

In this example, we will use NCW file. NMF file can also be used in the same way.

You can launch this file selection dialogue box by selecting main menu “File >> Import >>

Workfile(s)”. You can find these NCW files in the “06.Workfile” folder in the Sample Data

Folder (02.Org_Net_Tiny2)

Extracted_by_Gender and Age.ncw: Workfiles are assembled by Age and Gender

Extracted_by_Link_Attribute.ncw: Workfiles are assembled by Link Attribute

Extracted_by_Team.ncw: Workfiles are assembled by Team

88
 III.Tasks

Select the “Extracted_by_Team.ncw“ file, select Workfiles to import into NetMiner, and
Click OK. You can see the imported Workfiles as follows in the “Workfile Tree” area. You can

check the extracted Workfiles by Team.

Now let’s import the “Extracted_by_Gender and Age.ncw” file by selecting main menu
“File >> Import >> Workfile(s)”. You can see the following Workfiles imported into NetMiner

by checking the Workfile Tree area. The red lined box shows the Tree Structure from the
imported file.

In the same way, you can import the “Extracted_by_Link_Attribute.ncw “file.

89
Using NetMiner

 Exporting Workfile

You can export the Workfiles in the Workfile Tree for which you selected as a compressed file

with NMF extension. Tree Structure in the Workfile Tree represents data transformation of the

original Workfile, and it shows the process of data analysis history. So the compressed file

(NMF) can be useful for communication among the members who need to share the analysis

process. For example, it can be attached to email for communication.

Prior to understanding this exporting example, you may need to read “Importing Workfile” part

first. That is because we are going to practice with the imported Workfiles in the previous

“Importing Workfile” section. Launch the Workfile selection dialogue box for exporting as a

Workfile by selecting main menu “File >> Export >> Workfile(s)”. Select the

“02.Org_Net_Tiny2” item and the Workfiles concerning the Gender and Age as follows, and

then Click OK.

Write the name of the file to export and click OK. You can check the saved file in the selected

folder.

90
 III.Tasks

3. Creating Data

In this topic, creating data in NetMiner is covered. To create data in NetMiner, you need to

create Data Item in the first step. Then you can insert nodes and links in each created data item.

 Creating New Items

Data Items in NetMiner is one of the elements of Data Structure of NetMiner expressed in
Current Workfile area representing Main Nodeset, 1-mode Network data, Sub Nodeset and 2-

mode Network data.

You can select one of the three types of data item to create as follows by selecting “Data >>

Create New Item”. Since there can be only one Main Nodeset in one Workfile, Main Nodeset

Data Item cannot be created additionally in a Workfile. Other three types of Data Item can be

created in a Workfile, and multiple items for each type can be created.

• Creating Sub Nodeset Item

Select main menu “Data >> Create New Item >> Sub Nodeset” to launch a dialogue box for

creating Sub Nodeset Item as follows. Write the name of item as “My SubNodeset”.

You can check the created Sub Nodeset Item at Current Workfile area as follows.

91
Using NetMiner

You can recognize that the created data item is empty if you double click the data item you

created in the right side of the panel. Multiple Sub Nodeset Items can be created in a Workfile.

Let’s create one more Sub Nodeset Item. Create another Sub Nodeset Item as “My Sub

Nodeset2” by selecting main menu “Data >> Create New Item >> Sub Nodeset”.

Now we can insert nodes and attribute in the created Sub Nodesets. Refer to the topics about

“Inserting Nodes, Attributes and Links” in the following sections.

• Creating 1-mode Network Item

Now we are going to create 1-mode Network Data Item. Select main menu “Data >> Create

New Item >> 1-mode Network” for launching dialogue box for creating new item as follows.

Enter the title of 1-mode Network data and select the options regarding the direction and
multiple links. If you check the “Undirected Relation” check box, the created data cannot

represent directional information. If you check the “Allow Multiple Links” check box, the

created data can now express multiple links in a Edge List View. And now the data cannot be

edited in Matrix View anymore, since the matrix view cannot represent the multiple links.

Details about Multiple Links are covered in “Concepts – Data Structure – Representation of a
Network”. In this example, we will not select both of these options. Please click OK.

92
 III.Tasks

You can see the created item in the Current Workfile area as follow.

You can create multiple 1-mode Network Data Items. Create another 1-mode Network Data

Item as “My 1-mode Network Data2” by selecting main menu “Data >> Create New Item >> 1-

mode Network” as above.

But the created Data Items are empty. Now you can insert nodes and links in these Data Items.

Refer to the topic “Inserting Nodes, Attributes and Links”.

• Creating 2-mode Network Item

Now we are going to create 2-mode Network Data Item by selecting main menu “Data >>

Create New Item >> 2-mode Network”. In the dialogue box, you need to select the “Sub

Nodeset” for the 2-mode Network data item you are creating now. 2-mode Network data is

composed of two types of nodes: Main Nodeset and Sub Nodeset. Since Workfile can have
exactly one Main Nodeset, you cannot help but to select Sub Nodeset for comprising 2-mode
Network Data. In the “Nodeset” pull-down menu, you can select Sub Nodeset, and the two Sub

93
Using NetMiner

Nodesets have been already created. Please select “My SubNodeset”. If you check the “Allow

Multiple Links” check box, the created data can express multiple links in a Edge List View.

You cannot edit data in a Matrix View if you select this option. Details regarding Multiple

Links are covered in “Concepts – Data Structure – Representation of a Network”. In this time,
leave the Multiple Links option unchecked and click OK.

You can see the created 2-mode Network Data in the Current Workfile as follows. The created

2-mode Network Data is expanded as a sub item from the Sub Nodeset we selected.

Multiple 2-mode Network Data can be created in a Sub Nodeset. Let’s create a new 2-mode

Network Data as “My 2-mode Network Data2” for the “My SubNodeset”. And create a new 2-
mode Network Data as “My 2-mode Network Data3” for the “My SubNodeset2”. Please check

whether you have the same result with the below captured image of Current Workfile.

94
 III.Tasks

Now you can insert nodes, attributes and links to each of the created data item. Refer to the

topic: “Inserting Nodes, Attributes and Links”.

 Inserting Nodes, Attributes and Links

This topic covers “Inserting Nodes, Attributes and Links” in a created Data Item. You may need

to refer to the topic “Creating New Items” prior to this topic. We will insert nodes and attributes

in the created Main (Sub) Nodeset Data Item and links in the 1-mode (2-mode) Network Data

Item. We will use the example Project built from the topic “Creating New Items”. You can see

the created Data Items in the Current Workfile as follows.

• Inserting Main Nodes

You can check whether the data is empty by looking at the right side of the panel if you double

click the Main Nodeset Data Item which is noted as “Main” in the Current Workfile area at the

left side as shown below.

95
Using NetMiner

Select “Insert Node” in the mouse right click menu in the Row Label area (red lined area).

“Insert Node” menu is activated only in this area. You can see the dialogue box for inserting

nodes as follows. Write the “Node Name” as “My Node” You can insert multiple nodes at a

time if you select the “Number of Nodes”. We will insert 5 nodes at a time. Click OK.

You can see the inserted 5 nodes in the editing panel as follows.

96
 III.Tasks

You can insert Main Nodeset not only in the Main Nodeset Data Item but also in 1-mode

Network Data Item and 2-mode Network Data Item in the same way as above because Main

Nodeset is associated with other types of Data Item. You can check it by opening other Data

Items in the Current Workfile.

• Inserting Attribute Data

You can insert new attribute data in Main Nodeset or Sub Nodeset Data Item. We will insert

Main Node Attribute in this topic. You can insert Sub Node Attribute in the same way. .

You can launch the dialogue box for inserting attribute in the mouse right click menu at the

column label area as follows. Write the “Attribute Name” as “My Attribute” and select the

“Attribute Type” as Number. And Click OK.

97
Using NetMiner

You can see the inserted attribute column filled with “-999,999” which represents “Missing

Value”. For example, if the inserted attribute represents the age of each node, you can fill each

cell of the column with number as follows.

The supported data types in NetMiner are Number, Text, Date&Time and Time. You can
select one of these types when creating an attribute column.

• Inserting Links

There are two ways for inserting links in 1-mode (or 2-mode) Network data item in NetMiner.

One is to fill the cell in Matrix View with the weight of the link to be inserted, the other is to

select source node and target node to be linked in Edge List View. We will give an example
about inserting links in a 1-mode Network data. Inserting links in a 2-mode Network data can

be done in the same way.

- Matrix View

In Matrix View, you can insert links easily just by filling cells with weight of the links to be

inserted as follows. You can select Matrix View using Inner Tab at the bottom of the data
editing panel.

98
 III.Tasks

- Edge List View

You can see the 1-mode (or 2-mode) Network data in the editing panel at the right side of the

program in Edge List View as follows. You can select Edge List View using Inner Tab at the

bottom of the data editing panel. In Edge List View, you can insert a link by selecting source
node and target node to be linked.

Select “Insert Link” in the mouse right click menu in the row label area as follows. You can see

the dialogue box for inserting a link. In this dialogue box, you can select source node and target

node and the weight of the link to be inserted. You can check the inserted link at the last row of

the 1-mode Network data in Edge List View. If there are many nodes, it may be very difficult
to select node. You can use “Matching” test box for searching nodes matching with what you

wrote in the box.

99
Using NetMiner

If you try to insert a link with the nodes which are already linked, you will see the alarming

message as follows. If you have selected option for allowing “Multiple Link” at the dialogue
box when you having created 1-mode Network data item before, you can insert this kind of

“Multiple Link”. You need to refer to “Creating New Items” for more details.

100
 III.Tasks

4. Importing Data File

 Methods for Importing Data File

There are three ways for launching dialogue box for importing data file.

• Using “File >> Import”

You can import data file according to the file format to be imported. Select main menu “File

>> Import”, you can see the available file formats as follows.

You can select one of the file formats in the red lined box as above. Detailed options for each

dialogue box are covered in the following sections for each dialogue box.

• Using “Data >> Import Data Item”

You can import data file according to the data type to be imported. Select main menu “Data

>>Import Data Item”, you can see the available data type as follows. You can import only

Text/Excel file format if you import data file in this manner. Detailed options for each dialogue
box are covered in the following sections.

101
Using NetMiner

• Using Mouse Right Button Menu in Current Workfile

You can import data file according to the data type and file format to be imported. You can

select the data type by selecting the data item in Current Workfile using mouse left click and

select the file format using mouse right click menu as follows. Only the available file formats

are activated in the mouse right click menu that is matching to the data type you selected using

mouse left click. For example, if you selected 2-mode Network data item, StOCNET file format

will not be activated in the mouse right click menu. Detailed options for each dialogue box are

covered in the following sections.

 Importing Text/Excel File

Text/Excel file is generally used file format for importing data. All the options for both Text and

102
 III.Tasks

Excel file import dialogue box are same except the pull-down menu for selecting sheet in Excel

file. So, the Text and Excel file format Import dialogue box can be explained in this topic at one

time. We will use Excel data for example. The supported text file formats are TXT and CSV.

Refer to “Methods for Importing Data” about the ways for launching the dialogue box for

importing Text/Excel file. In this topic we will use main menu for launching the dialogue box:

“File >> Import >> Excel File”. Screenshot and short descriptions about the dialogue box are as

follows.

- Input File: You can launch the file selection dialogue box by clicking “Browse” button.
Excel file importing dialogue has pull-down menu for selecting sheet.

- File Preview: You can see some part of the file to be imported if you have selected file in
“Input File” area.

103
Using NetMiner

- Separator: In case of text file, you need to select separator for text.

- Text Qualifier: In case of text file, you need to select a qualifier that represents the text.

- Headers: You need to select Row/Column Header option.

- Data Type: You need to select the type of data. There are 4 types of data in NetMiner: Main
Nodeset, Sub Nodeset, 1-mode Network Data, 2-mode Network Data. You can select

“More” button for selecting detailed option.

- Target Workfile: You can select Target Workfile as “Current” or “New”. If you select
“Current”, the imported data will be added as a 1-mode Network data in the Current

Workfile. If you select “New”, the imported data will be contained in a new Workfile.

- Data Integrity Option: This option is activated when you selected “Current” as Target
Workfile. If you select “Insert newly discovered nodes”, new nodes in the imported data

will be added to the Main Nodeset of Current Workfile and the Main Nodeset will be

expanded. If you select “Discard newly discovered nodes”, new nodes in the imported data

will be discarded and the Main Nodeset will be maintained.

From now on, we will import the source data step by step using the sample data in the

following sample data folder: “02.Org_Net_Tiny2”. Sample data is described in “Introduction

– Sample Data Description”. Create a new Blank Project for data importing examples.

Quick Import: 1-mode Network Data

You can easily import 1-mode Network data. Select main menu “File >> Import >> Excel

File” for launching dialogue box for importing data. Click “Browse” button and select Excel

file in the file selection dialogue box in the following folder:

02.1-mode Network\01.Edge List\ Cooperation_unweighted.xls.

You can see some part of the file in File Preview. From File Preview, this file is 1-mode

Network data represented as Edge List type. It has two columns: source node and target node

and has no weight. CY001, CY037 etc. express the employee ID. This file represents the

cooperation relationship among the employees. This file does not have headers. So leave the
Header option unchecked. Select Data Type as Edge List. Leave the other options as default

104
 III.Tasks

and click OK.

You can see the import 1-mode Network data item in Current Workfile as follows. The default

name of the imported 1-mode Network data item uses the name of the Excel file and the sheet

name. You can see the content of the imported 1-mode Network data item at the right side of

the Current Workfile by double clicking the data item as follows.

If you double click the Main Nodeset data item, you can see the 34 nodes contained in the

imported Network data. These nodes are extracted from the imported 1-mode Network data
automatically. So these nodes have at least one link.

105
Using NetMiner

As illustrated above, it is easy to import 1-mode Network data. But if you import Main Nodeset

data prior to importing 1-mode Network data, you can control the Nodeset integrity.

Quick Import: 2-mode Network Data

You can import 2-mode Network data easily. Create a new Workfile for this example (File >>

New >> Workfile). Launch the dialogue box for importing data by selecting main menu “File

>> Import >> Excel File”. Click “Browse” button and select “Interest with coworkers.xls”

file in the sub folder in sample data folder as follows:

“04.2-mode Network\02.Matrix” in sample data folder.

If you open this Excel file, you can check a 2-mode Network data represented as a Matrix type
as follows. This data represents the relationship between the employees and the interest they

enjoy. The first column is Main Nodeset and the first row is Sub Nodeset. Now we are going to

import this file as a 2-mode Network data in NetMiner.

106
 III.Tasks

You can check some part of this data in “File Preview”. Select both of the Headers option. You

need to select Data Type as 2-mode Network represented as Matrix type. Leave other options as

default and click OK.

Double click the red lined 2-mode Network data item in Current Workfile and you can check
the content of data item at the right side.

107
Using NetMiner

You can check the Main Nodeset and Sub Nodeset by double clicking “Main” and “Interest

with coworkers_with_coworkers”. These Nodeset are extracted from the imported 2-mode

Network data automatically.

This method is useful if you need to analyze and visualize only one 2-mode Network data

quickly. You can add attribute data of Main Nodeset and Sub Nodeset to the imported data.

From now on, we will import Nodeset data prior to importing network data for comprising

systemic DataSet.

Importing Main Nodeset

Create a new Workfile for this example and name it as “Import Practice” (File >> New >>

Workfile”). Launch the dialogue box for importing Excel file by selecting main menu “File >>

Import >> Excel File”. Click “Browse” button and select the Excel file in the following folder :

Sample Data Folder: 01.Main Nodeset\ Employee_second.xls

This file represents attributes of 39 employees as Team, Gender etc. Open this file and check

it. Select Column Header check box. Select Data Type as Main Nodeset. Leave other options

set to the default selection.

108
 III.Tasks

Click “More” button for launching dialogue box for more advanced options. “Specify Node ID”

is for selecting Node ID column. If Node ID column is not in the first column, you need to

select the Node ID column in this option. Node Attributes shows the name and type of

attributes. You can change the name and type. Leave the options set to default selection and

click OK.

You can check the imported Main Nodeset data. Double click the Main Nodeset data item

109
Using NetMiner

named “Main” in Current Workfile. You can see the contents of data item at the right side of

this program called “Data Editing Panel”. Thirty nine employees and their attributes are
represented. And you can also check that new Workfile is expanded in Workfile Tree. Workfile

expansion occurs for every data transformation. In this case, Main Nodeset data importing

makes the new Workfile expansion.

The name of Workfile in this case is very long and complex. Default name of new Workfile is

composed of the file and sheet name to be imported. Let’s change it as “Main Nodeset”. Click

F2 key or use mouse right click menu “Rename”. And the parent Workfile in the Tree Structure

can be deleted as follows.

Importing 1-mode Network

We have imported 1-mode Network data in the previous “Quick Import” section. Now we are

going to import 1-mode Network data into Current Workfile which contains Main Nodeset data.
Let’s look at the sample data to import in this example. Open the following sub folder in

110
 III.Tasks

sample data folder: 02.1-mode Network.

You can see four sub folders: 01.Edge List, 02.Matrix, 03.Linked List, 04.Other Example.

We have made sample data in each of this folder for importing practice. They are classified by

data type. 1-mode Network data can be represented as three types: Edge List, Matrix and

Linked List.

① Edge List

You can see three Excel files in “01.Edge List” folder. Open each of these files. Screenshots of

them are as follows.

The simplest type of 1-mode Network data may be composed of just 2 columns: source node

and target node as you see at the left side of the pictures above. Its file name is

“Cooperation_unweighted.xls”. At the middle of these pictures, you can see “friendship_

weighted.xls” which is composed of three columns: source node, target node and weight.

At the right side of these pictures, you can see “knowledge_weighted_with_link_attribute.xls ”

file which has attribute of link as Category, Meeting Time and Feedback Score. Link

attribute is attached to each link between two nodes, it is not attached to each node. For

example, the “Category” which represents why the source node contacts the target node cannot

be attached to any single node. It is an attribute to describe the “Contact” between to nodes.

Full descriptions about “Link Attribute” are covered in “Concepts – Data Structure”.

Launch the dialogue box for importing Excel File by selecting main menu “File >> Import >>

111
Using NetMiner

Excel File”. Select “Cooperation_unweighted.xls” file by clicking “Browse” button.

There are no headers. So leave the Headers option unchecked. Select 1-mode Network as Data

Type and select Edge List as representation type. Select “Current” as Target Workfile. Full

description about Target Workfile will be covered in “Selecting Target Workfile” in the

following topic in this section.

You can check the imported 1-mode Network data in Current Workfile. Double click the 1-

mode Network data item (red lined box), you can see the contents of the data item at the editing

panel as follows.

112
 III.Tasks

Let’s rename the 1-mode Network data. It is too long and complex. Rename it as just

“Cooperation”.

Now we are going to import “friendship_weighted.xls” file that has weight column. This file

will be imported to Current Workfile that has already Main Nodeset and 1-mode Network data

(Cooperation). Launch the dialogue box for importing Excel File by selecting main menu “File

>> Import >> Excel File”. Select “friendship_weighted.xls” file by clicking “Browse” button.

Select options as follow and click OK.

You can check the imported data as follows.

113
Using NetMiner

You can check there are two 1-mode Network data in Current Workfile. Let’s rename the

imported data as “Friendship” as we have renamed the first 1-mode Network data as

“Cooperation”.

Now we are going to import “knowledge_weighted_with_link_attribute.xls” file that has

attributes of link. This file will be imported to Current Workfile that has already Main Nodeset

and two 1-mode Network data (Cooperation, Friendship). Launch the dialogue box for

importing Excel File by selecting main menu “File >> Import >> Excel File”. Select

“knowledge_weighted_with_link_attribute.xls” file by clicking “Browse” button. Select

options as follow.

114
 III.Tasks

Click “More” button at 1-mode Network area for launching dialogue box for detail option.

Direction is for selecting if the data is directed or not. Multiple Link is for handling Multiple

Link. If you select “Allow Multiple Link”, Multiple Link can be represented in Edge List View.

If you select “Merge Multiple Link”, Multiple Link is merged according to the Merge Option.

Link Columns is for selecting Source, Target and Weight Column. Link Attributes is for

handling attribute of links. You can change the name and type of attributes. Leave all the

options set to default selection and Click OK.

You can check the imported data as follows. We have imported third 1-mode Network data.

Now the Current Workfile contains three 1-mode Network data. This is because we have

selected “Current” as Target Workfile. Double click the imported 1-mode Network data, you
can see the content of this data item. If you see this data in Edge List View, you can check the

attribute of link. Let’s change the name of the imported data item as “Knowledge”.

115
Using NetMiner

② Matrix

Now we are going to import 1-mode Network data represented as Matrix type. Prior to

importing, open the file directly in the following folder:

“02.1-mode Network\02.Matrix\trust_matrix.xls”

You can see that source Nodeset and target Nodeset are same as follows.

Launch the dialogue box for importing Excel File by selecting main menu “File >> Import >>

Excel File”. Select “trust_matrix.xls” file by clicking “Browse” button.

Select both the Header options as checked. Select “1-mode Network data” as Data Type and

select “Matrix” as representation type. Select “Current” as Target Workfile. Leave other options

set to default and click OK.

116
 III.Tasks

You can see a new 1-mode Network data item in Current Workfile. Double click the new data

item (red lined box in the following picture), you can see the content of this data item in data

editing panel at the right side. Change the name of the data item as “Trust” in short.

117
Using NetMiner

③ Linked List

Now we are going to import 1-mode Network data represented as Linked List type. Prior to

importing, open the file directly in the following folder:

“02.1-mode Network\03.Linked List \ influence_linked_list.xls”

You can see that source Nodeset and target Nodeset are same as follows. The first column of

this file represents source node and the other columns represent target nodes. Note that each

source node has different number of target nodes.

Launch the dialogue box for importing Excel File by selecting main menu “File >> Import >>

Excel File”. Select “influence_linked_list.xls” file by clicking “Browse” button.

Leave both of the Header options unchecked. Select “1-mode Network data” as Data Type and

select “Lined List” as representation type. Select “Current” as Target Workfile. Leave other

options set to default and click OK.

118
 III.Tasks

Click “More” button for detail options. Check the Rank Option. Rank Option is to give order to

target nodes. If it is not checked, all the target nodes will have same weight. If it is checked,

target nodes will have weight in increase order from left to right. If you are to give weight in

decrease order from left to right, you can use main menu “Transform >> Value >> Reverse”.

Details can be referred to the “Module Reference”.

You can check a new data item is added in the Current Workfile as follow.

Change the name of the imported 1-mode Network data as “Influence”.

119
Using NetMiner

We have imported 5 Excel files into Current Workfile so far. There were three types of 1-mode

Network: Edge List, Matrix, and Linked List.

Importing Sub Nodeset

Sub Nodeset is used for comprising 2-mode Network data which requires two types of Nodeset:

Main Nodeset and Sub Nodeset. You can see three Sub Nodeset Excel files in the following

folder: “03.Sub Nodeset”

Interests.xls file represents the interests of employees. Clubs.xls file represents the clubs with

which the employees are affiliated. This file contains attributes of each clubs as Building and

Room. Movie.xls file represents the movies to which the employees give score. This file

contains attribute of each movie as Director, Category etc. These Sub Nodeset files will be used

for comprising 2-mode Network data.

First we are going to import Sub Nodeset one by one, and then import 2-mode Network data for

each of the Sub Nodeset.

120
 III.Tasks

Launch the dialogue box for importing Excel File by selecting main menu “File >> Import >>

Excel File”. Select “Interests.xls” file by clicking “Browse” button.

Leave both of the Header options unchecked. Select “Sub Nodeset” as Data Type. Leave Target

Nodeset set to default selection. The default name of Target Nodeset is made of the name of file

and sheet of the Excel file. Select “Current” as Target Workfile. Leave other options set to

default and click OK.

You can see a new Sub Nodeset data item in Current Workfile as follows. Double click the new

data item (red lined box in the following picture), you can see the content of this data item in
data editing panel at the right side. Change the name of the data item as “Interests”.

121
Using NetMiner

Now we are going to import “Clubs.xls”. Launch the dialogue box for importing Excel File by

selecting main menu “File >> Import >> Excel File”. Select “Clubs.xls” file by clicking

“Browse” button. Select the Column Header option checked. Select “Sub Nodeset” as Data

Type. Leave Target Nodeset set to default selection: Clubs_Sheet1. If you select “Interests” as

Target Nodeset, the imported Sub Nodeset data will be contained in the “Interests” data item.

We need to create a new data item and import data into this new data item, so we need to select

new Target Nodeset: Clubs_Sheet1. Select “Current” as Target Workfile. Leave other options

set to default and click OK.

You can see a new Sub Nodeset data item in Current Workfile as follows. Double click the new

122
 III.Tasks

data item (red lined box in the following picture), you can see the content of this data item in

data editing panel at the right side. Change the name of the data item as “Clubs”.

Now we are going to import “Movie Titles.xls”. Launch the dialogue box for importing Excel

File by selecting main menu “File >> Import >> Excel File”. Select “Movie Titles.xls” file by

clicking “Browse” button. Select the Column Header option checked. Select “Sub Nodeset” as

Data Type. Leave Target Nodeset set to default selection: Movie Titles_Sheet1. We need to

create a new data item and import data into this new data item, so we need to select new Target

Nodeset: Movie Titles_Sheet1. Select “Current” as Target Workfile. Leave other options set to

default. Click “More” button for additional options.

Specify Node ID is for selecting Node ID column. If Node ID column is not in the first column,

123
Using NetMiner

you need to select the Node ID column in this option. Node Attributes shows the name and type

of attributes. You can change the name and type. Leave the options set to default selection and

click OK.

You can see a new Sub Nodeset data item in Current Workfile as follows. When double clicking

the new data item (red lined box in the following picture), you can see the content of this data

item in data editing panel at the right side. Change the name of the data item as “Movie Titles”.

So far we have imported three Excel files of Sub Nodeset. Now we are going to import 2-mode

Network data for each of this Sub Nodeset.

124
 III.Tasks

Importing 2-mode Network

We have practiced importing 2-mode Network data in “Quick Import” topic. In that case, there

were no Main Nodeset and Sub Nodeset. So the Main Nodeset and Sub Nodeset were made of

the extracted Nodeset from the imported 2-mode Network data. In this topic, we are going to

import 2-mode Network data into Current Workfile which contains Main Nodeset and Sub

Nodeset. If there are multiple Sub Nodesets, you need to select one of them for the 2-mode

Network data to import.

Let’s look at the sample data to be imported in this example. Open the following sub folder in

sample data folder: 04.2-mode Network.

You can see three sub folders: 01.Edge List, 02.Matrix, 03.Linked List.

We have made sample data in each of these folders for importing practice. Those are classified

by data representation type. 2-mode Network data can be represented as three types: Edge List,

Matrix and Linked List.

① Edge List

You can see Excel file in “01.Edge List” folder. Open this file (“Movie Socre.xls”). Screenshot
of is as follows.

125
Using NetMiner

This file is composed of three columns: source node (Main Node), target node (Sub Node) and

Weight. Values of each cell in source node column represent the Node ID of employees and that

of target node column represents the Movie Titles to which the employees give score. The third

column represents the score the employees give to the move titles at second column.

Now we are going to import this 2-mode Network data into Current Workfile for the Sub

Nodeset “Movie Titles” which is already imported.

Launch the dialogue box for importing Excel File by selecting main menu “File >> Import >>

Excel File”. Select “Movie Score.xls” file by clicking “Browse” button. Select the Column

Header option checked. Select “2-mode Network” as Data Type and select Edge List as

representation type. Note that you need to select “Movie Titles” as Target Nodeset in the pull-
down menu. You can see four Sub Nodesets in the pull-down menu: Interests, Clubs, Movie

Titles and Movie Score_Score. Three of them are the Sub Nodesets we imported in the prior

steps. The last one is Sub Nodeset included in the 2-mode Network data to be imported right

now. If you select this last one, 2-mode Network data will be imported with creating a new Sub

Nodeset. We are going to import this 2-mode Network data for the one of the already imported

three Sub Nodesets. So we need to select one of them in the pull-down menu. Select “Current”

as Target Workfile. Leave other options set to default. Click “More” button for detail options.

126
 III.Tasks

You can see a new 2-mode Network data item for the selected Sub Nodeset in Current Workfile

as follows. After double-clicking the new data item (red lined box in the following picture), you

can see the content of this data item in data editing panel at the right side. Change the name of

the data item as “Movie Score”.

② Matrix

Now we are going to import 2-mode Network data represented as Matrix type. Prior to
importing, open the file directly in the following folder:

127
Using NetMiner

“04.2-mode Network\02.Matrix\”

You can see two Excel files named “Interest with coworkers.xls” and “Interest with

friends.xls”. The first column represents Main Nodeset and the first row represents Sub

Nodeset. Two files are composed of the same Main Nodeset and Sub Nodeset.

Launch the dialogue box for importing Excel File by selecting main menu “File >> Import >>

Excel File”. Select “Interest with coworkers.xls” file by clicking “Browse” button.

Select both of the Header options checked. Select “2-mode Network data” as Data Type and

select “Matrix” as representation type. Select “Interests” as Target Nodeset in the pull-down

menu.

Select “Current” as Target Workfile. Leave other options set to default and click OK.

128
 III.Tasks

You can see a new 2-mode Network data item for the selected Sub Nodeset in Current Workfile

as follows. Double click the new data item (red lined box in the following picture), you can see

the content of this data item in data editing panel at the right side. Change the name of the data

item as “Interests with Coworkers”.

Now in the same way, we are going to import “Interests with friends.xls” file. Leave both of

the Headers option unchecked. Leave other options set as was in the case of importing

“Interests with friends.xls” file and click OK.

129
Using NetMiner

You can see a new 2-mode Network data item for the selected Sub Nodeset in Current Workfile

as follows. Note that two 2-mode Network data item is in a Sub Nodeset “Interests”. Change

the name of the data item as “Interests with friends”.

③ Linked List

Now we are going to import 2-mode Network data represented as Linked List type. Prior to

importing, open the file directly in the following sample data folder:

“04.2-mode Network\03.Linked List \ Club Affiliation.xls”

The first column of this file represents source nodes (Main Nodes: Employees) and the other
columns represents target nodes (Sub Nodes: Clubs). Note that each source node has different

number of target nodes as in the case where one employee can affiliate multiple clubs.

130
 III.Tasks

Launch the dialogue box for importing Excel File by selecting main menu “File >> Import >>

Excel File”. Select “Club Affiliation.xls” file by clicking “Browse” button.

Leave both of the Headers Options unchecked. Select “2-mode Network data” as Data Type

and select “Linked List” as representation type. Select “Clubs” as Target Nodeset in the pull-

down menu. Select “Current” as Target Workfile. Leave other options set to default and click

OK.

Click “More” button for detail options. Rank Option is to give order to target nodes. If it is not

131
Using NetMiner

checked, all the target nodes will have same weight. If it is checked, target nodes will have

weight in increase order from left to right. Club affiliation does not need to represent the order.

So leave the Rank Option unchecked.

You can see a new 2-mode Network data item for the selected Sub Nodeset in Current Workfile

as follows. Double click the new data item (red lined box in the following picture), you can see

the content of this data item in data editing panel at the right side. Change the name of the data

item as “Club Affiliation”.

So far we have imported four 2-mode Network data Excel files into Current Workfile. There

were three types of 2-mode Network: Edge List, Matrix, and Linked List. Now we have

completed to compose a DataSet.

While performing all the importing process, the Workfile Tree may be very complex as follows.
Delete all the Workfiles except the last one, since the last Workfile contains all the imported

132
 III.Tasks

data in it. All the expansion in Tree Structure represents data transformation made so far.

We will change the name of the last Workfile as “Import Practice” as follows.

A DataSet in a Workfile can contain a Main Nodeset, multiple 1-mode Network data, multiple
Sub Nodeset, multiple 2-mode Network data for each Sub Nodeset as you have experience

through all the importing process. Full explanation about data structure of NetMiner is covered

in “Concepts – Data Structure”.

133
Using NetMiner

Selecting Target Workfile

Target Workfile option is for selecting the Target Workfile in which the imported data is

contained. If you select “Current”, the imported data will be contained as a new data item in

Current Workfile as you have seen in the previous topics. If you select the “New”, the imported

data will be contained in a newly created Workfile, which contains only one data to be being

imported.

Selecting “Current” as Target Workfile

Launch the dialogue box for importing Excel File by selecting main menu “File >> Import >>

Excel File”. Click “Browse” button and select “trust_matrix.xls” file in the following sample

data folder: 02.1-mode Network\02.Matrix\

Select both of the Headers Options checked. Select “1-mode Network data” as Data Type and

select “Matrix” as representation type. Select “Current” as Target Workfile. Leave other options

set to default and click OK

You can see a new 1-mode Network data item in Current Workfile as follows as we have

134
 III.Tasks

selected “Current” as Target Workfile.

Selecting “New” as Target Workfile

Now we are going to import same file. Select “New” as Target Nodeset. Select all the other

options set as same with the previous case. Write down the name of new Workfile as “New

Workfile Testing” and click OK.

You can see the new 1-mode Network data item in a newly created Workfile as follows when

135
Using NetMiner

we have selected “New” as Target Workfile.

Generally, to comprise a DataSet which composed of Main Nodeset, Sub Nodeset, 1-mode

Network Data and 2-mode Network Data you need to select “Current” as Target Workfile.

Handling Data Integrity Option

Data Integrity Option is for selecting whether or not to include new nodes while importing new

data into Current Workfile. This option is activated when “Current” is selected as Target

Workfile.

Let’s see the source data file for importing practice in this example in the following folder:

02.1-mode Network\04.Other Example\friendship2_weighted.xls

This file is 1-mode Network data. It has a new node that is not included in the 39 Main Nodeset

in the Current Workfile. We are going to import this file.

Launch the dialogue box for importing Excel File by selecting main menu “File >> Import >>

Excel File”. Click “Browse” button and select “friendship2_weighted.xls” file in the

following sample data folder: 02.1-mode Network\04.Other Example.

Select Column Headers Options checked. Select “1-mode Network data” as Data Type and

136
 III.Tasks

select “Edge List” as representation type. Select “Current” as Target Workfile. Select “Insert

newly discovered nodes” as “Data Integrity Option”.

You can see a new 1-mode Network data item in Current Workfile as follows.

Now double click the Main Nodeset data item named “Main”, you can see the content of the

selected data item. At the bottom of the Main Nodeset data, you can see a new node that is

inserted to Current Workfile while importing the “friendship2_weighted.xls” file. Because

this node does not have Main Node attribute value, cell values of attribute of Main Node is
filled with “Missing Value” as you see.

137
Using NetMiner

If you are to exclude new nodes while importing data into Current Workfile, you need to select

“Discard newly discovered Nodes” as Data Integrity Option.

Handling Multiple Link

Multiple Link represents links that are composed of same source node and target node. Multiple

Links can be represented as Edge List type of 1-mode (or 2-mode) Network data. Since

Multiple Link cannot be represented as Matrix type, in NetMiner 1-mode (or 2-mode) Network

data with Multiple Link can only be represented as Edge List View in Data Editing Panel. You
can handle Multiple Link in the dialogue box with more options by selecting “More” button.

You can allow Multiple Links or merge those by summation while importing process.

First, let’s check the Multiple Link in sample source file in the following sample data folder:

02.1-mode Network\01.Edge List\friendship_weighted.xls

Open this file and look at the bottom of this data. There you will see an example of the Multiple
Link as follows.

Launch the dialogue box for importing Excel File by selecting main menu “File >> Import >>

Excel File”. Click “Browse” button and select “friendship_weighted.xls” file.

138
 III.Tasks

Select Column Headers Options checked. Select “1-mode Network data” as Data Type and

select “Edge List” as representation type. Select “Current” as Target Workfile. Select “Insert

newly discovered nodes” as “Data Integrity Option”.

Click “More” button for selecting detail options. Select “Allow Multiple Link” as Multiple Link

option as follows. Leave other options set as default selection and click OK.

You can see a new 1-mode Network data item in Current Workfile as follows. Double click the
new data item; you can see the content of the selected data item. As you see, this data item can

139
Using NetMiner

only be seen in Edge List View because there is Multiple Link in this network data. You can

find these pairs of Multiple Link: (CY269, CY167, 2), (CY269, CY167, 5).

If you select “Merge” in the Multiple Link option while importing the same file, the pairs of

Multiple Link will be merged as follows: (CY269, CY167, 7). After merging Multiple Links,

both Matrix View and Edge List will be available as the content representation of data item.

 Importing Multiple Text/Excel Files

This option allows you to import multiple data files of various types at once.

We will create a new Workfile named “Multiple Import Practice”. Launch the dialogue box by

selecting main menu “File >> Import >> Multiple Text/Excel File(s)”. Click “Add” button for

adding a row in which you can select data file and options.

There are three columns: File, Type and Option. You can see a button for launching file

selection dialogue box if you click the cell represented as “<Select Data File here>” in File
column. Select “Employee_second.xls” in the following sample data folder: 01.Main

140
 III.Tasks

Nodeset. Select “Main Node (+ Attribute)” in Type pull-down menu. If you click the cell in

Option column, you can see a dialogue box for detail options as follows. Check the “Include

Header Row” option. Select the Node ID column in “Select Columns” option and click Close.

Now Main Nodeset is ready to be imported. But in this example, multiple data will be imported

all at once including 1-mode Network, Sub Nodeset and 2-mode Network data in this importing

process.

Add three more rows in the dialogue box by importing below-listed files:

- 1-mode Network data: trust_matrix.xls (Matrix type)

- Sub Nodeset: Movie Titles.xls

- 2-mode Network data: Movie Score.xls

Select each data file and select data type for each file as follows.

141
Using NetMiner

Descriptions for option setting for each 3 files are as follows

Trust_matrix.xls: Select “Full Matrix” as Link Type and check “Row Label” and “Column

Label”.

Movie Titles.xls: Select “Include Header Row” check box. Select “Titles” as Node Column.

142
 III.Tasks

Movie Score.xls: Select “Include Header Row” as Header Format. Select “Score” as Weight

Column.

Now all the option settings are finished. Click Next button for next step to import data as

follows. You can see some information about 1-mode Network and 2-mode Network data. If

you are to import the data file as an undirected Network data, select “No” in Directed column.

Multiple Link column shows if the data file has Multiple Link or not.

143
Using NetMiner

Click Next button for next step. You can see some information about Main Nodeset and Sub

Nodeset data. If you are to change the type of the selected attribute, select type for that attribute

in Type column.

Click Next button to proceed. You can see a User Interface for setting Sub Nodeset

classification. This step is for selecting Sub Nodeset for selected 2-mode Network data. You
can decide the Sub Nodeset to which the selected 2-mode Network data belongs.

144
 III.Tasks

Select two items in the left panel using mouse with holding CTRL key, and click right arrow

button at the center of this panel as shown below.

You can see that the selected items at left panel moved to right panel as follows.

Click next for next step. You can see the short report about the importing result. It shows how

many nodes and links are imported as follows. Click Finish button to complete the importing

process.

145
Using NetMiner

You can see new items in Current Workfile as follow.

 Importing NTF File

NTF file is a file format that was used in prior version of NetMiner. NTF file is editable using
external editor. Open the sample NTF file in the following sample data folder:

146
 III.Tasks

02. Org_Net_Tiny2\Import\ 02.Org_Net_Tiny2.ntf

You can import NTF file into NetMiner 4.

Create a new Project for this example.

Launch the dialogue box for importing NTF file by selecting “File >> Import >> NTF File”.

NTF file does not contain a data, but it is a file with a set of Data, so that it cannot be directly

added to Current Workfile. A new Workfile must be created to contain the NTF file to be

imported. Other options for each data are not activated.

Click Browse button for selecting “02.Org_Net_Tiny2.ntf “file and click OK. You can see that

the DataSet in the NTF file is import in Current Workfile as follows.

147
Using NetMiner

 Importing DL File

DL file is a file format that is used in UCINet. It is editable in external editor. NetMiner 4 can

import DL file. Only number type is available in DL file. And it cannot contain the 1-mode

Network and 2-mode Network at one time because only one Sub Nodeset can be contained in a

DL file.

Importing DL file containing 1-mode Network data

Open the sample DL file in the following sample data folder:

02. Org_Net_Tiny2\07.UCINet_DL\1-mode Networks.dl

This file contains five 1-mode Network data.

Launch the dialogue box for importing DL file by selecting “File >> Import >> DL File”. Select
Browse button and select the DL file. Select “New” as Target Workfile. Select “1-mode

148
 III.Tasks

Network” as Data Type and click OK.

You can see a message box as follows.

You can see new data items in Current Workfile as follows. Five 1-mode Network data is

imported into NetMiner 4. New Workfile is added because we have selected “New” as Target

Workfile.

149
Using NetMiner

Importing DL file containing 2-mode Network data

Open the sample DL file in the following sample data folder:

02. Org_Net_Tiny2\07. UCINet_DL\2-mode Network_Club Affiliation.dl

This file contains a 2-mode Network data.

Launch the dialogue box for importing DL file by selecting “File >> Import >> DL File”. Select

Browse button and select the DL file. Select “Current” as Target Workfile. Select “2-mode

Network” as Data Type and leave Target set to default selection.

You can see new 2-mode Network data items in Current Workfile as follows.

In this way, NetMiner DataSet can be composed of multiple DL files.

150
 III.Tasks

 Importing Pajek File

.NET, .VEC are file extensions of Pajek file format. It is editable in external editor. NetMiner

can import these kinds of Pajek files. These file formats can contain only number types. NET

file is to represent 1-mode (2-mode) Network data, while VEC file is to represent a vector.

Net file is composed of two parts: vertex (node) list first and after that edge list. “*ARC”

represents that this data has direction. “*EDGES” represents that it has no direction.

In case of 2-mode Network data, at the first row it must be designated the number of Main

Nodes in the VERTICES list. For example, “*VERTICES 56, 39” means that there are 39 Main

Nodes in total 56 nodes in VERTICES list. The rest nodes are Sub Nodes.

Importing Net file containing 1-mode Network data

151
Using NetMiner

Open the sample NET file in the following sample data folder:
02. Org_Net_Tiny2 \08.Pajek\1-mode Network_L_Cooperation.net

This file contains a 1-mode Network data.

Launch the dialogue box for importing NET file by selecting “File >> Import >> Pajek File”.

Select Browse button and select the file above. Select “New” as Target Workfile. Select “1-mode

Network” as Data Type and click OK.

You can see a message box as follows.

You can see new 1-mode Network data item in Current Workfile as follows. New Workfile is

added because we have selected “New” as Target Workfile.

152
 III.Tasks

Importing Net file containing 2-mode Network data

Open the sample NET file in the following sample data folder:
02. Org_Net_Tiny2\08.Pajek\2-mode Network_L_Club Affiliation.net

This file contains a 2-mode Network data. It represents the relationship between the employees

and the clubs to which they are affiliated.

Launch the dialogue box for importing NET file by selecting “File >> Import >> Pajek File”.

Select Browse button and select the file above. Select “Current” as Target Workfile. Select “2-

mode Network” as Data Type and click OK.

You can see new 2-mode Network data item in Current Workfile as follows.

In this way, NetMiner DataSet can be composed of multiple NET files.

153
Using NetMiner

VEC file can also be imported into NetMiner 4 as an attribute data.

 Importing StOCNET

NetMiner can import text file formats (TXT, DAT) used in StOCNET. StOCNET text files can

contain only number type values and represent only 1-mode Network data with Matrix type.

Open the sample StOCNET file in the following sample data folder:
02. Org_Net_Tiny2\09. StOCNET\1-mode Network_L_Cooperation.dat

This file contains a 1-mode Network data as follows.

Launch the dialogue box for importing StOCNET file by selecting “File >> Import >>
StOCNET File”. Select Browse button and select the file above. Select “New” as Target

Workfile. Select “1-mode Network” as Data Type and click OK.

You can see a message box as follows and after that you can see a new 1-mode Network data

154
 III.Tasks

item in Current Workfile as follows.

 Importing GML File

NetMiner can import GML file. Open the sample GML file in the following sample data folder:

02. Org_Net_Tiny2\ ? GML\karate\karate.gml

You can see the content of this file as follow.

Select main menu “File >> Import >> GML File” for importing GML file. There is no

dialogue box for importing GML file unlike other formats which required dialogue box in the

previous examples.

You can see a new 1-mode Network data item in Current Workfile as follows.

155
Using NetMiner

156
 III.Tasks

5. Importing DB Tables

“Importing DB Tables” module, in File menu, is not available for Student-Education License.
The supported DB systems are Oracle, MS-SQL, MySQL and MS Access database over ODBC

connection. The method to import DB Table is almost the same with that of data files. But the

available type of 1-mode (or 2-mode) network data is only Edge List type. Matrix and Linked

List type are not supported in importing DB Tables.

Launch the dialogue box for importing DB Tables by selecting main menu “File >> Import DB

Tables.” Click the Connect button to launch a dialogue box for setting connection information.

You can see DB Connect dialogue box as follows.

157
Using NetMiner

You can make a new profile about the information of the DB system as the Server IP, DB Name,

User Name and Password. You can save the connection profile. Click OK, you can see the

tables in the connected DB system as follows. At the top of this dialogue box, you can see the

Server IP and the DB Name. You can see the tables that are contained in this DB in the pull-

down menu.

You can select the type of data to be imported, as we did in the previous section in which the

methods to import data files were discussed. Only Edge List type is allowed in the 1-mode (or

2-mode) Network data type. Full descriptions regarding selecting options in detail are covered

in the previous section “Importing Data”.

In this section, we are going to import 1-mode Network data “Friendship_weighted” table as an
example, First, you need to select “Friendship_weighted” in the Table pull-down menu.

158
 III.Tasks

You can see some part of the data in Table Preview. And you can check some information about

the selected table in Table Information as the column name and the attributes etc. You need to

select 1-mode Network in Data Type option and click OK to import table.

159
Using NetMiner

You can see the imported DB Table in NetMiner as follows.

160
 III.Tasks

6. Exporting Data

 Methods for Exporting Data

“Exporting Data” is to save data in NetMiner Data Structure as an external file formats. There

are three ways to launch dialogue box for exporting data.

- Using “File >> Export”

You can export data file according to the file format to be exported. Select main menu “File >>

Export”, you can see the available file formats as follows.

You can select one of the file formats in the red lined box as above. Detailed options for each

dialogue box are covered in the following sections for each dialogue box.

- Using “Data >> Export Data Item”

You can export data file according to the data type. Select main menu “Data >>Export>> Data

Item”, you can see the available data type as follows. You can export only as Text/Excel file

format if you export data using this way. Detailed options for each dialogue box are covered in
the following sections for each dialogue box.

161
Using NetMiner

- Using Mouse Right Button Menu in Current Workfile

You can export data according to the data type and file format to be exported. You can select

the data item to export in Current Workfile using mouse left click and select the file format

using mouse right click menu as follows. You can export multiple data item at one time if you

selected multiple data item (using mouse left click and CTRL, SHIFT key) in Current Workfile.

Detailed options for each dialogue box are covered in the following sections.

We will use “2.Org_Net_Tiny2.nmf” file as an example for exporting data from the next topic.

162
 III.Tasks

 Exporting to Text File

The exported text file can contain only one data item, such as a Main Nodeset data, a Sub

Nodeset data and a 1-mode (or 2-mode) Network data etc.

Launch the dialogue box for data item to be exported by selecting main menu “File >> Export

>> Text File” as shown below, where you can select data item(s) to be exported. We will select

all the items using Select All check box. Click OK.

You can see a dialogue box for selecting options for exporting data as follows.

163
Using NetMiner

- Output File: You can select a folder to save exported file by clicking Browse button. Prefix
is for giving prefix to the name of the exported files. The name of each Sheet is composed

of following structures: [Prefix] + [_L (or_N)] + [Data Item Name]. Prefix “L” and “N” are

abbreviations of Linkset (“L”) and Nodeset (“N”), respectively. Prefix “L” is used when the

exported data type is 1-mode (or 2-mode) Network data. Prefix “N” is given when the

exported data type is Main Nodeset or Sub Nodeset. CSV and TXT formats are available as

the File extension.

- Separator: You can select separator for each value in text file.

- Text Qualifier: You can select qualifier for text type value. You can apply the selected
qualifier to Date&Time type also by checking Apply to Date and Time.

- Nodeset Option: You can select whether or not to represent Column Header. You can
select whether or not to export attributes of nodes with the Nodeset.

- Network Option: You can select the representation type of network data: Matrix type or
Edge List type. If you export as Matrix type, you can select whether or not to include

Row/Column Headers. And you need to select Merge Option if Multiple Link is included in

the exported data. If you export as Edge List type, you can select whether or not to include

Column Header, Link Weight and Link Attribute in the exported file.

For this example, select a folder for saving data. Leave the Prefix set to default value. Select

“CSV” as File extension. Select “Comma” as Separator and “Quotation Mark” as Text Qualifier.

Select both of the Nodeset Options and select “Export to EdgeList” as Network Option. And
click OK you can see a message as follows.

You can see that the CSV files are saved for each data item in the selected folder as follows.

164
 III.Tasks

If you open “02.Org_Net_Tiny2_L_knowledge.csv” file for example, you can see the

content of it as follows. You can check that text type data is qualified by quotation marks and

attributes of link are included.

 Exporting to Excel File

An Excel file can contain multiple data in each Sheet. NetMiner 4 takes advantage of this
feature so that you can save as a single Excel file for selected multiple data items in each Sheet.

Launch the dialogue box for data item selection for exporting by selecting main menu “File >>

Export >> Excel File”. You can see a dialogue box as follows. You can select data item(s) to

export. We will select all the items using Select All check box. Click OK.

165
Using NetMiner

You can see a dialogue box for selecting options for exporting data as follows.

- Output File: You can select a folder and file name to save exported file by clicking Browse
button.

- Nodeset Option: You can select whether or not to represent Column Header. You can
select whether or not to export attributes of nodes with the Nodeset.

- Network Option: You can select the representation type of network data: Matrix type or
Edge List type. If you export as Matrix type, you can select whether or not to include

Row/Column Headers. And you need to select Merge Option if Multiple Link is included in

the exported data. If you export as Edge List type, you can select whether or not to include

Column Header, Link Weight and Link Attribute in the exported file.

Select a folder and file name to save exported file by clicking Browse button. Select both of the

166
 III.Tasks

Nodeset Options and select “Export to EdgeList” as Network Option. And click OK, you can

see a message box as follows.

You can see that an Excel file is saved containing each data item in Sheets as follows. The

name of each Sheet is composed of following structures: [_L (or_N)] + [Data Item Name].

Prefix “L” and “N” are abbreviations of Linkset (“L”) and Nodeset (“N”), respectively. Prefix

“L” is used when the exported data type is 1-mode (or 2-mode) Network data. Prefix “N” is

given when the exported data type is Main Nodeset or Sub Nodeset.

 Exporting to NTF File

You can export the selected data item as NTF file format, which is old format used by the prior

version of NetMiner 4. NTF file format can contain multiple data items in a file, but it is

limited to represent only number type values. So make sure that you have only number type
values to be exported. Values with other type such as TEXT will be excluded in the exporting

167
Using NetMiner

process.

Launch the dialogue box for data item selection for exporting by selecting main menu “File >>

Export >> NTF File” as shown below. Upon a dialogue box, you can select data item(s) to be

exported. The concept of Sub Nodeset is not defined in NTF format so that Sub Nodeset data

item cannot be selected for exporting. We will select all the items using Select All check box.

Click OK.

You can see a dialogue box for selecting options for exporting data as follows.

- Output File: You can select a folder and file name to save exported file by clicking Browse
button.

- Nodeset Option: Since NTF file format has fixed type for Nodeset representation, this
option is not available.

- Network Option: You can select the representation type of network data: Matrix type or
Edge List type. If you export as Matrix type, Row/Column Headers will not be included in

168
 III.Tasks

the exported NTF file. And you need to select Merge Option if Multiple Link is included in

the exported data. If you export as Edge List type, you can select whether or not to include,

Link Weight and in the exported file. NTF cannot represent the attribute of Link.

You can see a message box that shows which attribute is excluded in exporting process. These

attributes are excluded because the values of these attributes are not Number type.

You can see that an NTF file is saved as follows. You can open this file in the Prior version of

NetMiner.

 Exporting to DL File

You can save data items in NetMiner 4 as DL file whose format is used in UCINet. It is editable

169
Using NetMiner

in external editor. Please note that it should not contain the 1-mode Network and 2-mode

Network at one time because only one Sub Nodeset can be contained in a DL file. Only number

type is allowed in DL file, and values with other types such as TEXT will be excluded in the

exporting process.

Launch the dialogue box for data item selection for exporting by selecting main menu “File >>

Export >> DL File”. You can select data item(s) to export as in the dialogue box below. We

will select all the items using Select All check box. Click OK.

You can see a dialogue box for selecting options for exporting data as follows.

- Output File: You can select a folder to save exported file by clicking Browse button. Prefix
is for giving prefix to the name of the exported files. The name of each Sheet is composed

of following structures: [Prefix] + [_L (or_N)] + [Data Item Name]. Prefix “L” and “N” are
abbreviations of Linkset (“L”) and Nodeset (“N”), respectively. Prefix “L” is used when the

170
 III.Tasks

exported data type is 1-mode (or 2-mode) Network data. Prefix “N” is given when the

exported data type is Main Nodeset or Sub Nodeset. DL is available as the File extension.

- Nodeset Option: Since DL file format has fixed type for representing Nodeset, this option
is not available.

- Network Option: You can select the representation type of network data: Matrix type or
Edge List type. If you export as Matrix type, Row/Column Headers will not be included in

the exported DL file. And you need to select Merge Option if Multiple Link is included in

the exported data. If you export as Edge List type, you can select whether or not to include,

Link Weight and in the exported file.

A message box shows which attributes are being excluded from exporting process. These

attributes are excluded because their attributes are not in Number type.

You can see that the DL files are saved in the selected folder as follows.

All the 1-mode Network data are contained in a single file. But each of the 2-mode Network

data is contained in its corresponding file, separately. There is also a file containing attribute of

Main Nodeset.

 Exporting to Pajek File

171
Using NetMiner

You can save data items in NetMiner 4 as Pajek NET file or VEC file. The exported NET file

contains only one Network data (1-mode or 2-mode) and the exported VEC file contains only

one attribute vector.

Launch the dialogue box for data item selection for exporting by selecting main menu “File >>

Export >> Pajek File”. You can see a dialogue box as follows. You can select data item(s) to

export. We will select all the items using Select All check box. Click OK.

You can see a dialogue box for selecting options for exporting data as follows.

- Output File: You can select a folder to save exported file by clicking Browse button. Prefix
is for giving prefix to the name of the exported files. The name of each Sheet is composed
of following structures: [Prefix] + [_L (or_N)] + [Data Item Name]. Prefix “L” and “N” are

172
 III.Tasks

abbreviations of Linkset (“L”) and Nodeset (“N”), respectively. Prefix “L” is used when the

exported data type is 1-mode (or 2-mode) Network data. Prefix “N” is given when the

exported data type is Main Nodeset or Sub Nodeset.

- Nodeset Option: Since DL file format has fixed type for representation Nodeset, this
option is not available.

- Network Option: You can select the representation type of network data: Matrix type or
Edge List type. If you export as Matrix type, Row/Column Headers will not be included in

the exported DL file. And you need to select Merge Option if Multiple Link is included in

the exported data. If you export as Edge List type, you can select whether or not to include,

Link Weight and in the exported file.

You can see that the NET files and VECT files are saved for each data item in the selected

folder as follows.

 Exporting to StOCNET File

You can save data items in NetMiner 4 as StOCNET file. In this case Sub Nodeset and 2-mode

Network data cannot be exported. Only number type values can be exported and only one 1-

mode Network data item can be contained in a file.

Launch the dialogue box for selecting data item to be exported by navigating main menu “File

>> Export >> StOCNET File”. You can see a dialogue box as shown below. You can select
data item(s) to be exported. You will find that only 1-mode Network data items are allowed. We

173
Using NetMiner

will select all the items using Select All check box. Click OK.

You can see a dialogue box for selecting options for exporting data as follows.

- Output File: You can select a folder to save exported file by clicking Browse button. Prefix
is for giving prefix to the name of the exported files. The name of the exported file is

composed of following elements: [Prefix] + [_L (or_N)] + [Data Item Name]. Prefix “L”

and “N” are abbreviations of Linkset (“L”) and Nodeset (“N”), respectively. Prefix “L” is

used when the exported data type is 1-mode (or 2-mode) Network data. Prefix “N” is given

when the exported data type is Main Nodeset or Sub Nodeset. CSV and TXT formats are

available as the File extension.

- Nodeset Option: This option is not available.

- Network Option: Only Matrix type is available. Row/Column Headers will not be included

174
 III.Tasks

in the exported file. And you need to select Merge Option if Multiple Link is included in the

exported data.

175
Using NetMiner

7. Editing Data

 Overview

Data editing in NetMiner occurs in a Data Editing Session for each data item. A Data Editing

Session is composed of two parts: Data Editing Panel (blue lined box in the center) and Data

Control Panel (green lined box in the right). Data Editing Session is activated when you double

click the data item in Current Workfile. You can activate multiple Data Editing Session as

follows.

 Actions in Data Editing Panel

• “View” of Editing Panel

The UI representation of Data Editing Panel depends on the type of data item. For example,

Data Editing Panel of Nodeset Data is as shown below. At the bottom of the Panel, notice that
the inner tab is named as Editor.

176
 III.Tasks

In case of Network data (1-mode Network or 2-mode Network), there are two inner tabs named

Matrix and Edge List at the bottom of the panel. The content of each Network data item will be

represented based upon its corresponding type: Matrix View and Edge List View in NetMiner.

In Matrix View, you can insert a link just by filling a cell with number that represents weight.

177
Using NetMiner

With small number of nodes, since entire matrix can be seen so that you can grasp the global

pattern of Network. Edge List View represents only links with weight, which consumes less

storage space than Matrix View. It can also represent Multiple Link and Link Attribute which

cannot be represented in Matrix View.

You can edit 1-mode Network data using Graph Editor by clicking “Open Graph Editor” button

in Control Panel. While Graph Editor is activated, you can see a new inner tab written “Map” at

the bottom of panel. This appearance is called Map View. Full description about using Graph

Editor is covered in “Editing Data Using Graph Editor”.

<Graph Editor Capture Image>

• Selecting

178
 III.Tasks

You can select the cells in Data Editing Panel in various ways. The color of selected cells

changes to blue as follows. The color of selected row and column label area changes to orange.

The descriptions about selection in row label area are as follows.

You can select all the data in Data Editing Panel if you choose the cell at the upper-left corner.

Or there are other options such as mouse right click menu or keyboard (CTRL + A) for

selecting all the data. For selecting a row, you need to select a cell in the first column with row

index of 1, 2, 3, 4 and so on. In the same way, you can select a column by selecting the first row

of the column.

Different mouse-right click menu will be triggered, based upon the area at which the click

occurs. For example, the menu in Row/Column Label area (gray colored) is different from the
menu in cell value area (white colored). To select mouse right click menu for inserting/deleting

179
Using NetMiner

row/column, you need to click mouse right button in Row/Column Label area with gray colored.

• Writing / Modifying

You can fill or modify the value in a cell just by clicking mouse left button or just press F2

function key. In the same way, you can edit the values in Row/Column area. In Matrix View,

filling a cell with some values means creating a link in the network.

In Edge List View, source node and target node (colored sky) is not editable. You can only edit

the weight column and attribute of link column. You need to use a dialogue box for inserting a

link in Edge List View. Details are covered in “Inserting/Deleting Link”.

• Copying / Pasting

You can cut/copy/paste values in Data Editing Panel. You can do these editing works in external

editing program like Excel. You can use mouse right click menu or generally used shortcut key

as CTRL+X,C,V.

When you are to paste data in Row Label area, you may encounter the message box, as shown

below, for data integrity check. Make sure that Row Labels should be unique.

You can copy data in Editing Panel to Excel and copy Excel data to NetMiner. Double click 1-

mode Network data item “Trust”. In Data Editing Panel, select all the data and copy that to

Excel.

180
 III.Tasks

You can see the pasted data in Excel as follows. It is far easier than using main menu “File >>

Export >> Excel File”.

Now we are going to copy this data in Excel to a newly created 1-mode Network data. Create a

new 1-mode Network data item as follows.

Change the name of this data item as “Editing Test”. Open this created data item in Data

Editing Panel by double clicking the data item.

181
Using NetMiner

Now we are going to copy the data in Excel to NetMiner. Select all the data in Excel and copy-

and-paste those in NetMiner as shown below. Click the red lined cell in NetMiner as shown

below and paste the data by CTRL+V. You can check that data is copied to NetMiner.

Alternatively, it is possible to paste data by selecting the red lined cell and choosing “Paste”

option. The Row/Column Label will be pasted at the first row/column, so it requires adding a

row/column to the current data. You can select whether or not to expand the current Nodeset.

We will select “Preserve,” then that first row and column are filled with missing value that

182
 III.Tasks

represents Row/Column Label.

You need to copy data in Excel without the row/column label as follows to avoid this problem.

You can copy and paste data between the data items in NetMiner. Select some area by mouse

dragging in “Trust” data and copy that to “Editing Test” data item. You can check it as follows.

* Shortcut for Editing data in Excel

You can use simply the “Open in Excel” in mouse
right click menu for editing data in Excel. Using this

fuction, you don’t have to copy the data in NetMiner

and paste it again in Excel sheet.

183
Using NetMiner

• Inserting/Deleting Node(s)

In Matrix View, values in Row Label area represent Main Nodes. If you are to insert
node(s), select “Insert Node” in the mouse right click menu at the Row Label area.

You can see a dialogue box for inserting node(s) as follows. You need to write the name of

node(s). You can select the position of inserted node(s). You can insert multiple nodes at one

time. Serial number will be attached to each name of multiple nodes.

You can check the inserted nodes as follows.

Since the inserted Main Node(s) may affect other data items, you need to check the integrity of
other data items after you inserted Main Node(s). For example, when you inserted Main

184
 III.Tasks

Node(s) in 1-mode Network data item, you can check the inserted Main Node(s) in Main

Nodeset data item, as well. In that case, you can see that the inserted Main Node(s) does not

have attribute of nodes as follows.

You can delete node(s) in the same way. Select node(s) to delete using mouse left click and

select “Delete Node” menu in the mouse right click menu. You can also delete multiple nodes

at one time.

You can also insert/delete nodes in the same way for Sub Nodeset. You cannot insert/delete

nodes in Edge List View.

• Inserting/Deleting Attribute

You can insert/delete attribute of nodes for both of Main Nodeset and Sub Nodeset. And you

can insert/delete attribute of links for both of 1-mode Network data and 2-mode Network data
in Edge List View. In Matrix View, you can add/modify/delete each value of the attribute of link,

but you cannot insert/delete the items of attribute.

If you are to insert attribute of Main Node, select “Insert Attribute” in the mouse right click

menu at Column Label area. You can see the dialogue box for inserting attribute as follows.

You can write down the name of attribute and select the type of the attribute. The available

types are Number, Text, Date&Time, Time.

185
Using NetMiner

If you have inserted an attribute item, you can see a new column, which is filled with System

Missing Value denoted as “-999,999” as shown below. You need to fill each cell with the values

for that attribute.

In the same way, you can delete an attribute item. Select “Delete Attribute” in the mouse right
click menu at the Column Label area.

In the same way, you can insert/delete attribute of link using mouse right click menu at the

Column Label area in Edge List View.

• Inserting / Deleting Link

Inserting link in Matrix View is very simple; you can insert link just by filling the empty cell in
the Matrix View. However, inserting link in Edge List View may not be obvious at first. First,

186
 III.Tasks

select “Insert Link” in the mouse right click menu at the Row Label area. You can see a

dialogue box for inserting link as follows. Then select each of the source node and target node

and the weight of this link.

 Actions in Data Control Panel

• Checking Network Information

You can check whether or not the selected Network data has Multiple Links and Direction. You

can check this information only for 1-mode (2-mode) Network data.

• Controlling Number Precision

You can select the number of decimal points to be represented by Number Precision. Data

Editing Panel in NetMiner can express to the sixth decimal points.

187
Using NetMiner

• Permutating Matrix

In Matrix View, you can change the order of Main Nodes in terms of the attribute of Main

Nodes. In case of 1-mode Network data, Row and Column order will change at the same time.

In case of 2-mode Network data, Row order will change.

• Choosing Link Attribute

You can select the attribute of link to be represented in Data Editing Panel in Matrix View. For

example, you can see the attributes of link in Edge List View if you select 1-mode Network data

item “Knowledge”. In Matrix View, default values that are represented in Data Editing Panel are

weight. You can change the values by selecting attribute of link in Choose Link Attribute in

Control Panel. Select “Meeting Time” in the pull-down menu, you can see the Time type values

in Data Editing Panel.

• Value Labeling

You can give a label to the original value. For example, it is useful to assign text type value to
the original value of number type, because numerical calculation is available with the original

188
 III.Tasks

numerical value while having the labels as text representation.

We will give labels to an attribute of Main Nodes. Let the numerical values in “Age” attribute

labeled as “Twenties” or “Thirties”. You can see the dialogue box for assigning label as follows

if you click Add button in Value Label Control Item.

Select “Age” in Attribute pull-down menu. Enter the text value to assign its label. You can see

the available label values in the selected attribute item by checking Show Values check box.

Double click the value to give a label, and the selected value will be represented in Value text

box. Write down the label to give to the selected value in Label text box. Click OK button, and
you can see that the selected value is given the label as follows.

• Missing Value Setting

You can designate the selected values in a selected attribute item as “Missing Value”. You can

189
Using NetMiner

see the dialogue box for missing value setting by clicking Add button in Missing Value Control

Item as follows. We will designate “17” in “Duration” attribute item as “Missing Value” for

example. Select “Duration” in Attribute pull-down menu. Write down “17” in Value text box.

You can see the available values by checking Show Values check box. Double click the value to

designate as “Missing Value”, you can see this value in Value text box. And click OK button.

You can see that the color of the designated value changes to red.

 Editing Data Using Graph Editor

• Launching Graph Editor

You can see a network map representing the selected 1-mode Network data item with gray

colored background in Data Editing Panel: a new View named Map. You can also check that

Display Control Panel is added to the Control Panel.

190
 III.Tasks

You can see a network map representing the selected 1-mode Network data item with gray

colored background in Data Editing Panel: a new View named Map. You can also check that

Display Control Panel is added to the Control Panel.

• Inserting / Deleting Nodes

To insert a node, click the gray colored area using mouse left button. You can see a new node

named “New Node (1)” as follows. Adding a new node can be done in this simple manner.

You can check the inserted new node in Main Nodeset data. Double click the Main Nodeset
data item in Current Workfile, you can see the inserted node at the bottom of the data as follows.

191
Using NetMiner

You can see that attributes of this inserted node are “Missing Value”.

To delete node(s), first select the nodes to delete and select “Delete Selected Nodes” in the

mouse right click menu or press DEL key in keyboard.

• Inserting / Deleting Link(s)

To insert a link, select source node and target node one by one. If you select a node on the

network map in Graph Editor, the color of the selected node will change to blue and this

selected node is designated as source node. If you select another node for designating as target

node, a link between the previously-selected blue colored node and the now-selected node will
be created. The weight of the new created link is 1 as default value. Select “Edit Link Attribute”

in the mouse right click menu to change the weight or the attribute of the selected link.

To delete link(s), select the link(s) to delete and select “Delete Selected Links” in mouse right

192
 III.Tasks

click menu or press DEL key in keyboard.

• Controlling Network Map Panel

You can zoom in (out) the network map in Graph Editor. You can see the Navigation Control

Item in Display Control Panel as follow. You can zoom in (out) the network map by using

sidebar and navigate in Preview Map. Click Fit to Screen button to fit the resized network map

to the current size of panel. Click Layout button to run the layout algorithm for drawing

network map again.

• Selecting Nodes and Links

You can select nodes and links using mouse left click on network map. The color of the selected

nodes and links will be changed to yellow as follows. You can select nodes and links in addition

to the previously-selected nodes and links using mouse dragging with SHIFT key. If you select

the previously-selected nodes and links using mouse dragging with ALT key, the selected nodes

and links will be excluded as deselected nodes and links. If you select nodes and links using

mouse dragging with CTRL key, the selection will be reversed: the previously -selected nodes

and links will be deselected and the pre-deselected nodes and links will be selected. You can
use “Select All” in mouse right click menu for selecting all the nodes and links.

193
Using NetMiner

• Modifying Node Name and Node Attribute Value

You can modify the name and the attribute value of a selected node in Graph Editor. But you

can not insert or delete attribute items. You can see a dialogue box for modifying values by

selecting Edit Node Attribute in the mouse right click menu at the selected node as follows. You

can change the values in the red lined box area.

194
 III.Tasks

• Modifying Link Weight and Link Attribute Value

You can change the weight and the attribute value of a selected link in Graph Editor. But you

can not insert or delete attribute items. You can see a dialogue box for modifying values by

selecting Edit Link Attribute in the mouse right click menu at the selected link as follows. You

can change the values in the red lined box area.

195
Using NetMiner

8. Transforming Data Using Matrix Calculator

 Overview

Matrix Calculator is a tool for transforming and manipulating data in Current Workfile using

command line interface as follows like a shell. You can launch Matrix Calculator by selecting

main menu “Tools >> Matrix Calculator”. User Interface of Matrix Calculator is composed of

two parts as follows: red lined box area and blue lined box area.

You can input command for manipulating data in Command Panel. You can see the

manipulation results in Command Panel. You can create a new variable in Command Panel.

You can add the created new variable to Current Workfile in Variables Panel. You can change

panels by selecting inner tabs at the bottom of the panel named Command and Variables.

196
 III.Tasks

The blue lined area is a panel for selecting data, functions and operators. You can see the

available data items in Current Workfile and supported functions and operators. You can input

these data, functions and operators in command line just by double clicking each of them.

These panels for selection are activated in Command Panel.

The Command Panel is divided into two parts: lower area for command input and upper area

for manipulation results. Although command can be received by keyboard directly, it will be

easier to use mouse for selecting data, functions and operators in the blue lined panel. Special

care should be taken when using command line especially for case-sensitiveness.

- Supported Data Types

Matrix Calculator supports four types of data as follows.

① Integer. ex) 3, 5, 325, 614

② Real Number ex) 3.4, 6.5, 7.1

③ String: Text type data. ex) "NetMiner", "Cyram"

④ Matrix: Matrix type data composed of Real Number type elements.

ex) 1 0 1

023

003

- Supported Functions

① getMatrixSum () : return a scalar as the sum of the elements in a selected matrix

② getRowSum () : return a column vector which is composed of the sum of the elements in a
row in a selected matrix

③ getColumnSum () : return a row vector which is composed of the sum of the elements in a
column in a selected matrix

④ getDiagonalVector() : return a column vector which is composed of the diagonal elements

in a selected square matrix

197
Using NetMiner

⑤ trace () : return a scalar as the sum of the diagonal elements in a selected square matrix

⑥ getRowVector() : return a column vector which is composed of the elements in a selected

row in a selected matrix

⑦ getColumnVector() : return a column vector which is composed of the elements in a

selected column in a selected matrix

⑧ transpose () : return a transposed matrix of a selected matrix

⑨ inverse () : return an inversed matrix of a selected matrix

⑩ dichotomize () : return an dichotomized matrix of a selected matrix (dichotomization

criteria = 0)

⑪ symmetrize () : return a symmetrized matrix of a selected matrix (symmetrization criteria

= Max.)

- Supported Operators

① + : addition

② - : subtraction

③ * : multiplication

④ ** :power

⑤ / : division

⑥ elemWiseMult(): elementwise multiplication

⑦ elemWiseDiv(): elementwise division

⑧ eleWisePow(): elementwise power

- Basic Operation

Matrix Calculator supports basic operations of Integer and Real Number type data. Write down

the expressions in the command line as follows. Matrix Calculator recognizes number with

decimal points as Real Number type. In this case, the returned result is also Real Number type.

Operations of text type data are also supported as follows.

198
 III.Tasks

Examples.

1+1=2

5/3=1

5/3.0=1.6667

(24+7)/3=10

(24+7)/3.0=10.333

“Cyram” + “ “ + “NetMiner” = “Cyram NetMiner”

 Retrieving Data From Current Workfile

You need to write down the name of function and variable in command line to retrieve data

from Current Workfile. But it is recommended to use Control Panel at the right side. Double

click the name of variable in the panel, you can see the selected variable with matching

functions for retrieving data is written in the command line.

- Main Node Attribute: _MAT.FromNM.getMainNodeAttribute("The Name of Attribute Data") -

199
Using NetMiner

Retrieving Main Node Attribute as a numerical data

ex. _MAT.FromNM.getMainNodeAttribute ("Duration")

- Main Node Attribute: _MAT.FromNM.getMainNodeAttribute("The Name of Attribute Data") -

Retrieving Main Node Attribute as its data type

ex. _MAT.FromNM.getMainNodeAttribute ("Duration")

- 1-mode Network: _MAT.FromNM.get1modeNetwork ("The Name of 1-mode Network Data")

ex. _MAT.FromNM.get1modeNetwork ("Trust")

- Sub Node Attribute: _MAT.FromNM.getSubNodeAttribute ("The Name of Sub Nodeset Data","

The Name of Attribute Data ") - - Retrieving Sub Node Attribute as a numerical data

ex. _MAT.FromNM.getSubNodeAttribute ("Movie Titles","Category")

- Sub Node Attribute: _MAT.FromNM.getSubNodeAttribute ("The Name of Sub Nodeset Data","

The Name of Attribute Data ") - Retrieving Sub Node Attribute as its data type

ex. _MAT.FromNM.getSubNodeAttribute ("Movie Titles","Category")

- 2-mode Network: _MAT.FromNM.get2modeNetwork ("The Name of Sub Nodeset Data","The

Name of 2-mode Network Data")

ex. _MAT.FromNM.get2modeNetwork ("Clubs","Club Affiliation")

For example, we will retrieve 1-mode Network data “Trust”. Double click the variable “Trust”

in 1-mode Network data item. You can see that the variable and matching function

“_MAT.FromNM.get1modeNetwork ("Trust")” is written in command line as follows. Press

Enter key, you can check the result of this command in the upper area of the command line.

200
 III.Tasks

 Creating a Variable

You can create a variable without declaration. You can create a new variable just by assign a

value to a new variable. The type of the created variable is decided according to the type of the

assigned value. Space, ‘ ’, is not supported for the name of a variable.

Examples.

Creating an Integer variable: a=5

Creating an Real Number variable: b=3.5

Creating a 1-mode Network data variable: c=_MAT.FromNM.get1modeNetwork ("Trust")

Creating a Main Nodeset data variable: d=_MAT.FromNM.getMainNodeAttribute ("Duration")

201
Using NetMiner

 Using Functions and Operators

You can use functions and operators in Matrix Calculator. Available functions and operators list

and the methods for using them in command line are covered in “Module Reference”. It is

recommended to use functions and operators by double clicking them in Control Panel. First,

select a function using mouse, then the cursor will be located at a position where a variable is

required. Then select a variable using mouse, resulting in that the selected variable with

matching function for retrieving data will be written in that position. You can select the

operators in the same way.

 Adding Calculated Matrix to Current Workfile

The newly-created variable containing the calculated results can be added to the Current

Workfile. You can see the created new variables in Variable Panel. If you check the Show

Values check box, you can see the elements contained in the selected variable. Using the

options in Add to DataSet area, you can add the created variables to Current Workfile. For

adding calculated matrix to Current Workfile, the number of row and column in the created

variable must be equal to the number of row and column in the Current Workfile. For example,

if the number of Main Nodeset in Current Workfile is n, the number of Main Nodes in the 1-

mode Network to be added to Current Workfile must be n. You need to select the target Sub
Nodeset if you are to add a variable to Sub Node attribute or 2-mode Network data.

 Examples

Calculating Weighted Sum of 1-mode Network Data and Adding it

to Current Workfile

Step1: Double click the 1-mode Network data “Trust” in 1-mode Network Control item. You
can see that _MAT.FromNM.get1modeNetwork ("Trust") is written in the command line.

202
 III.Tasks

Step2: Input weight value which is multiplied to the selected 1-mode Network data as follows.
_MAT.FromNM.get1modeNetwork ("Trust") * 3: Multiply by 3 for all the elements in 1-mode

Network data “Trust”

Step 3: In the same way, we will retrieve 1-mode Network data “Knowledge” and multiply
weight value 5 as follows.
_MAT.FromNM.get1modeNetwork ("Trust") * 3+_MAT.FromNM.get1modeNetwork ("knowledge")*5

Step 4: In the same way, input all the 1-mode Network data and weight for each data and sum
all of those as follows.
_MAT.FromNM.get1modeNetwork ("Trust")*3

+_MAT.FromNM.get1modeNetwork ("knowledge")*5

+_MAT.FromNM.get1modeNetwork ("Influence")*4

Step 5: Press Enter key, you can see the calculated results in the upper area as follows. But it is
not easy to recognize the full pattern of the matrix.

203
Using NetMiner

Step 6: Now we will create a new variable and assign the calculated result to this created
variable. Hit the up-arrow key to get to your previously used commands in history buffer. We

will assign the result of this expression to a new variable “new” as follows.

new = _MAT.FromNM.get1modeNetwork ("Trust")*3

+_MAT.FromNM.get1modeNetwork ("knowledge")*5

+_MAT.FromNM.get1modeNetwork ("Influence")*4

You can see the calculated results in the upper area of the command line.

Step 7: Now move to the Variable Panel by selecting the Variable Tab. Select the variable
“new” and press Add button for 1-mode Network data at the Add to DataSet area.

204
 III.Tasks

You can see a new 1-mode Network data item “new” in Current Workfile as follows.

In this way, you can add a new variable to data items in Current Workfile.

205
Using NetMiner

Extracting Column Vector from 2-mode Network Data and Adding

it to Main Node Attribute

Step 1: In this example, we will add a column of a 2-mode Network data to the attribute of
Main Nodeset. In 2-mode Network data “Score”, we will extract score vector for the movie

“The Godfather” and add it to the Main Nodeset. First, double click “Score” in the Control

Panel.

Step 2: Double click the function “getColumnVector”. This function is for extracting a selected
column vector (parameter) from a specific matrix. Select “1” as the parameter. You can see the

function at the command line.

Step 3: Press Enter key and you will see the extracted column vector as follows.

Step4: Now we will create a new variable and assign the calculated result to this new variable.
Hit the up-arrow key to get to your previous commands from the command-history list buffer.
We will assign the result of this expression to a new variable “score1” as shown below.

206
 III.Tasks

score1= _MAT.FromNM.get2modeNetwork("Movie Titles","Score").getColumnVector(1)

Step5: Now move to the Variable Panel by selecting the Variable Tab. Select the variable
“score1” and press Add button for Main Nodeset at the Add to DataSet area.

You can see the added column vector in the attribute of Main Nodeset in Current Workfile as

follows.

207
Using NetMiner

9. Using Output Tables Recursively

 Overview

The Output Tables generated in a session can be added to Current Workfile for another analysis

and visualization process. Reuse of the generated result as a data for another process is another

strong feature of NetMiner.

The generated output tables can be added as node attribute of Main Nodeset or as link attribute

of 1-mode(or 2-mode) Network data. It can also be added as a new 1-mode (or 2-mode)

Network data item according to the type of the output table. The generated output tables also

can be saved as a new Workfile containing multiple data items which came from the output

tables.

You can use main menu “Data – Add Output Table (or Save Output Tables)” or mouse right

click menu at the Output Table. The available type is activated in these menus.

 Adding as Main Node Attribute

Vector(s) in an Output Table can be added to Main Nodeset data as attribute of Main Nodeset.

Step1. Creating a Session by selecting main menu “Analyze >> Neighbor >> Degree”. Select
1-mode Network data “Trust” and press Run Process button for generating output. You can see

the outputs as follows.

208
 III.Tasks

The generated outputs are as follows: a Main Report ([R] Main), Two Output Tables ([T]

Degree, [T] Node Type) and a Network Map ([M] Spring). You can see the vector(s) in [T]

Degree, [T] Node Type. These vectors can be added to attribute of Main Nodeset.

Step 2. In Output Table [T] Degree, select main menu “Data >> Add Output Table >> Main
Node Attribute” as follows.

Step 3. You can see the dialogue box for selecting vector(s) to add to Main Node attribute. You
can see the available vector(s) in current Output Table. You can change the name of vector(s).

209
Using NetMiner

We will change each of them as “InD, OutD”. Click OK.

Step 4. Now let’s check the added attributes. In Current Workfile, double click Main Nodeset
data item, you can see the added attributes at the end of the data column as follows.

Now we will add “Node Type” vector in [T] Node Type to Main Node Attribute using mouse

right click menu. In the Output Table, select “Add to Workfile >> Node Attribute”, you can see

the dialogue box for selecting vector(s). You can add the selected vector to Main Node Attribute

in the same way as above.

210
 III.Tasks

 Adding as Link Attribute

If the type of data in Output Table is 1-mode (or 2-mode) Network, it can be added as link

attribute of 1-mode (or 2-mode) Network data in Current Workfile. Only the values of the

Output Table with 1-mode (or 2-mode) Network data type matching to the links of the 1-mode

(or 2-mode) Network data will be added.

Step1. Create a Link Connectivity Session by selecting main menu “Analyze >> Connection
>> Connectivity >> Link”. Select 1-mode Network data “Trust” and press Run Process button.

Select the Output Table ([T] Link Connectivity Matrix), and then you can see a matrix type data
representing the link connectivity value between two nodes in row and column. More detailed

description regarding Link Connectivity is covered in NetMiner Module Reference.

211
Using NetMiner

Step 2. Now we will add this data in Output Table as a link attribute of “Trust” 1-mode
Network data. In the selected Output Table, select the “Add to Workfile >> 1-mode Link

Attribute” in the mouse right button menu as follows.

You can see a dialogue box for selecting 1-mode Network data to add the output data as link

attribute as follows. Select 1-mode Network data. You can rename the link attribute data in

Attribute Name text box.

212
 III.Tasks

Step 3. You can see a new 1-mode Network link attribute data is added as follows.

213
Using NetMiner

 Adding as 1-mode Network Data

If the type of data in Output Table is 1-mode Network, it can be added as 1-mode Network data

item in Current Workfile.

Step1. Create a Node Connectivity Session by selecting main menu “Analyze >> Connection
>> Connectivity >> Node”. Select 1-mode Network data “Trust” and press Run Process button.

Select the Output Table ([T] Node Connectivity Matrix), you can see a matrix type data

representing the connectivity value between two nodes in row and column. Node Connectivity

is the number of nodes to be deleted for disconnecting the selected two nodes. Details are

covered in Module Reference.

Step 2. Now we will add this data in Output Table to 1-mode Network data item. In the selected
Output Table, select main menu “Data >> Add Output Table >> 1-mode Network”, you can see
a dialogue box for naming the 1-mode Network data as follows. Fill the text box and click OK.

214
 III.Tasks

Step 3. You can see a new 1-mode Network data in Current Workfile as follows. Double click
the data item and check its data in Data Editing Panel on the right.

 Adding as 2-mode Network Data

If the type of data in Output Table is 2-mode Network, you can add this data to Current

Workfile as a new 2-mode Network data for a new created Sub Nodeset.

Step1. Create Clique Session by navigating main menu “Analyze >> Cohesion >> Clique” and
select 1-mode Network “Trust”. Press Run Process button for generating Output. Selecting
Output Table [T] Clique Affiliation Matrix shows you a data with 2-mode Network type. This

215
Using NetMiner

output represents the Cliques in the selected 1-mode Network data and which Clique each node

belongs to.

Step 2. Select main menu “Data >> Add Output Table”, you can see that both of Main Node
Attribute and 2-mode Network are activated. You can add the data in current Output Table to

Main Node Attribute or 2-mode Network data. If you are to add the data to Main Node Attribute,

you need to select vector(s). Refer to the details about this topic in “Adding as Main Node

Attribute”. We will select 2-mode Network. You can see a dialogue box as follows.

Step 3. You can see a new 2-mode Network data item in Current Workfile as follows.

216
 III.Tasks

 Adding Multiple Tables

A Session sometimes generates multiple Output Tables, and those multiple sessions can be

activated for interaction with users. Moreover, it is possible to insert those multiple Output

Tables in multiple sessions into each data item at one time.

First, activate all the sessions used in the above examples: Degree, Node Connectivity, and
Clique.

Double click each session in Current Workfile. We will add the outputs in all the sessions. So

you need to press Run Process button for each of the sessions.

Step1. Select main menu “Data >> Add Output Table >> Multiple Tables”, you can see the
dialogue box for selecting Output Tables in the activated sessions. Select all the Output Tables

and click OK.

217
Using NetMiner

Step 2. Output Tables have been added to the Current Workfile and new data items will be
available in Current Workfile.

 Saving as a New Workfile

You cannot add data in Output Table to Current Workfile, when the number of the row of the

data in Output Table differs from the number of Main Nodes in Current Workfile. In this case,

you need to save the data in Output Table as a new Workfile.

Step 1. Create the Blockmodel session by navigating main menu “Analyze >> Position >>
Blockmodel (Conventional)”. Select 1-mode Network data “Trust” and you need to select
partition vector “Team”. Press Run Process button for generating output. Six Output Tables are

generated. But, as you can see, the number of rows in Output Tables is different from the

number of Main Nodeset in Current Workfile. In this example, the rows in the Output Tables

are the name of team we selected as a partition vector. So, the data in Output Tables cannot be

added to Current Workfile. You need to save the data in Output Tables as a new Workfile.

218
 III.Tasks

Step 2. After selecting main menu “Data >> Save Output Tables”, a new Workfile is expanded
from Current Workfile with the following message. Click OK, and then the newly saved

Workfile is created as Current Workfile.

Step 3. Each data in Output Tables in Blockmodel session is saved as data item in the new
Workfile where each data item matches its corresponding type of data in Output Tables.

219
Using NetMiner

220
 III.Tasks

10. Querying Data

 Overview

Querying is for extracting nodes and links from Current Workfile with specific condition. You

can create / save / edit / apply this condition for extracting nodes and links.

Generally, extracting Sub DataSet from a DataSet and comparing the extracted Sub DataSets

are important part of data analysis. Querying is for supporting these processes. It helps users to

extract nodes and links conveniently and to apply the extracted nodes and links to analysis and

visualization process.

QuerySet is the specific condition for extracting nodes and links. Note that QuerySet is not a
set of extracted nodes and links but the condition itself for extracting nodes and links. QuerySet

differs from Selection. QuerySet is the condition itself for extracting nodes and links, but

Selection is the set of nodes and links selected on the network map. Full explanations are
covered in “QuerySet vs. Selection” in this section.

You can create (or edit), save and apply QuerySet for analysis. You can create (or edit)

QuerySet in Query session by selecting main menu “Tools >> Query Composer”. Then you can

save the created QuerySet. You can see the saved QuerySets in QuerySet items in Current
Workfile. You can apply the saved QuerySets to the selected analysis/visualization module in

Filter Control item. Using Filter for applying QuerySet and Selection is covered in “Process

Handling – Handling Filter”.

In this section, we are going to explain how to create and save a QuerySet.

221
Using NetMiner

 User Interface for Querying

Select main menu “Tools >> Query Composer”, you can see the user interface of Query

Session. Double click the data item in the right side (red lined box).

 QuerySet Status: You can see the data structure in Current Workfile and check how many
nodes and links will be extracted by the Query Sentence for each data item in QuerySet.

The numbers in parentheses are representing the extracted nodes (or links) and the total

nodes (or links). The numbers in parentheses represent number of nodes for Nodeset data

item and represent number of links for network data item. You can check (and edit) the

Query Sentence for each data item by double clicking the data item in Query Status.

 Target: The name of data item which you selected in QuerySet Status for creating (or
editing) Query Sentence.

 Shortcut: You can input Query Sentence in Query area using your keyboard. But it is

222
 III.Tasks

difficult to input the name of variable, operators and value without error when just typing

it. And you need to remember its available values for a selected variable. In Short Cut,

you can select variables and values for selected variable and select operators using mouse

click. It will help you to build Query Sentence without error.

 Query: You can build Query Sentence in Query area. You can use your keyboard or
mouse click in Short Cut. Detail descriptions on grammar for making out Query Sentence

are available in “Module Reference”.

 Log: You can just try the Query Sentence for checking how many nodes and links will be
extracted by hitting Try button. The results will be shown in Log area. If you applied the

Query Sentence to QuerySet Status by hitting Apply button, you can check it in this area

also.

 Reset button: Recover the default status.

 Load button: You can load the saved QuerySet for editing (or just checking).

 Save button: You can save the QuerySet.

 Run button: You can extract nodes and links meeting the current QuerySet and save it as a
new Workfile.

 Writing Query Sentence for Each Data Item

• Query Sentence for Extracting Nodeset

You can extract Nodeset with specific condition from Main Nodeset or Sub Nodeset. For

example, we will build Query Sentence for extracting Nodeset from Main Nodeset. You can do

this process for Sub Nodeset in the same way.

Example 1 - Simple Sentence: "Gender" == "male"

Step 1. Create Query Session by selecting main menu “Tools >> Query Composer”. Double
click Main Nodeset data item “Employee” in QuerySet Status in order to set Main Nodeset data
item as Target data item for building Query Sentence.

223
Using NetMiner

Step 2. We are going to extract nodes with the condition: “Gender is male”. First, double click
“Gender”. Second, click operator “= =”. Third, select “male” in the Gender pull-down menu.

You can see the Query Sentence in Query area as follows: "Gender" == "male"

But it is still possible to type in the query, but it is less error-prone to use Query UI for helping

you to write query statement easily and correctly.

224
 III.Tasks

Step 3. You can just ‘test’ to check how many nodes will be extracted by this Query Sentence
by hitting Try button. You can see the result in Log area as follows. 31 nodes will be extracted

from total 39 nodes in this data item as you can see in Log area.

Now you can apply this Query Sentence to QuerySet Status by hitting Apply button. Full

descriptions are covered in “Applying Query Sentence to QuerySet Status”.

Example 2 – Complex Sentence: "Team" =="Finance" and "Age" >= 30.0

Step 1: As we did in example 1, create Query Session and set Main Nodeset as Target data item
for building Query Sentence. You can refer to the descriptions in step 1 in example 1.

Step2: Select variable, operator and value as follows.

 Double click variable “Team”

 Click operator “==”

 Select “Finance” in variable “Team” pull-down menu

 Click operator “and”

 Double click variable “Age”

 Click operator “>=”

 Select “30” in variable “Age” pull-down menu

225
Using NetMiner

Step 3: You can just ‘test’ to check how many nodes will be extracted by this Query Sentence
by hitting Try button. You can see the result in Log area: 8 nodes will be extracted from total 39

nodes in this data item in Log area.

Now you can apply this Query Sentence to QuerySet Status by hitting Apply button. Full

descriptions are covered in “Applying Query Sentence to QuerySet Status”.

Node ID in Main Nodeset can also be used in Query Sentence. You can give specific condition

to node ID. You need to use special variable “@LABEL” for representing node ID in Query
Sentence.

• Query Sentence for Extracting Linkset

You can extract Linkset with specific condition from 1-mode (or 2-mode) Network data. A

condition for a link can be given to the weight or the attribute of link. In addition to this, two

incident nodes of a link can be given specific condition about the attribute of each node as a

condition for the link between these two nodes. All these operations for giving specific

conditions to links can be achieved by Querying. First, create Query Session by selecting main

menu “Tools >> Query Composer”.

Example 1 – Simple Sentence: @WEIGHT >= 3.0

Step 1: Select a 1-mode Network data “Knowledge” as Target data item when querying. You
can see the available items for building Query Sentence in Short Cut like Source Node, Target

Node, Weight and various attributes of link as follows.

226
 III.Tasks

Step 2: Select variable, operator and value as follows.

① Double click variable “@WEIGHT”

② Click operator “>=”

③ Select “3.0” in variable “@WEIGHT” pull-down menu

Step 3: You can just ‘test’ to check how many links will be extracted by this Query Sentence by
hitting Try button. You can see the result in Log area as follows. 74 links will be extracted from

total 84 links in this data item as you can see in Log area.

Now you can apply this Query Sentence to QuerySet Status by hitting Apply button. Full

descriptions are covered in “Applying Query Sentence to QuerySet Status”.

Example 2 – Complex Sentence: @WEIGHT >= 3.0 and @SOURCE."Position" == "Team

Member” and @TARGET."Position" == "Team Manager" and "Category" == "Decision Making"

227
Using NetMiner

Step 1: Select a 1-mode Network data “Knowledge” which contains attributes of link for setting
it as Target data item for querying.

Step 2: We will give a complex condition to “Knowledge” Network for extracting Linkset as
follows: (Weight >= 3) and (the purpose of knowledge flow is “Decision making”) and (Source

node is “Team Member”) and (Target node is “Team Manager”). Select variable, operator and

value as follows.

① Double click variable “@WEIGHT”

② Click operator “>=”

③ Select “3.0” in variable “@WEIGHT” pull-down menu

④ Click operator “and”

⑤ Double click variable “Category”

⑥ Click operator “==”

⑦ Select “Decision Making” in variable “Category” pull-down menu

⑧ Click operator “and”

⑨ Double click variable “@SOURCE”

⑩ Select “Position“ in variable “@SOURCE” pull-down menu

⑪ Select “Team Member“ in variable “Position” pull-down menu

⑫ Click operator “and”

⑬ Double click variable “@TARGET”

⑭ Select “Position“ in variable “@TARGET” pull-down menu

⑮ Select “Team Manager“ in variable “Position” pull-down menu

Step 3: You can just ‘test’ to check how many links will be extracted by this Query Sentence by
hitting Try button. You can see the result in Log area where 22 links will be extracted from total

84 links in this data item as you can see in Log area.

Now you can apply this Query Sentence to QuerySet Status by hitting Apply button. Full
descriptions are covered in “Applying Query Sentence to QuerySet Status”.

228
 III.Tasks

 Applying Query Sentence to QuerySet Status

You can apply Query Sentence to QuerySet Status by hitting Apply button. You can check it in

Log area. QuerySet Status represents cumulated results of the Query Sentences. You can check

how many nodes and links will be extracted by the parentheses attached to each data item.

We will use Example 2 for extracting Nodeset and Example 1 for extracting Linkset. You can

refer to these examples in previous topics. The condition for that is ["Team" =="Finance” and

"Age" >= 30.0] and the condition for this is [@WEIGHT >= 3.0]. Now we are going to apply

this Query Sentence to QuerySet Status.

Step 1. Hit the Apply button for applying following Query Sentence that we made out in
example 1 in the previous topic for extracting Nodeset: ["Team" =="Finance" and "Age" >=

30.0]

You can see that the number of nodes in the parentheses in Main Nodeset data item changes as

follows. You can check how many nodes will be extracted.

Step 2. Hit the Apply button for applying following Query Sentence that we built in example 2
in the previous topic for extracting Linkset: [@WEIGHT >= 3.0]

229
Using NetMiner

You can see that the number of links in the parentheses in 1-mode Network data “knowledge”

item changes as follows. You can check how many links will be extracted.

You need to check that the two Query Sentences are applied to QuerySet Status cumulatively.

Query Sentences are applied to QuerySet Status. Now you can save it as a QuerySet or Run it

for extracting nodes and links and saving it as a new Workfile.

 Saving QuerySet

You can save the QuerySet, which can be found in QuerySet item in Current Workfile. This

stored QuerySet in Filter Control item in Process Panel can be used for each analysis and

visualization session.

We will re-use the QuerySet example, which we used in previous topic “Applying Query

Sentence to QuerySet Status” for this topic.

Launch the dialogue box for saving QuerySet by hitting save button as follow. Input the name

of the QuerySet to be saved. It will be helpful if you input the conditions of the QuerySet as the

name. We inputted “My QuerySet 1 ”. Click OK button.

230
 III.Tasks

You can see a new item in QuerySet in Current Workfile as follows.

 Loading QuerySet

You can load saved QuerySet for editing or adding new Query Sentence. Launch the dialogue

box for selecting QuerySet by hitting Load button as follows.

231
Using NetMiner

You can load the selected QuerySet in Current Workfile directly using mouse right click menu.

Select a QuerySet to load using mouse left click and select Load in mouse right click menu as

follows.

You can check the Query Sentence for each data item by double clicking data item in QuerySet

Status. If you want to recover the default Query Sentence, click Restore Current button.

 Using Saved QuerySet in Filter

You can use QuerySet in Filter Control item in Process Panel for each analysis and

visualization session. Detailed descriptions are covered in “Handling Filter”.

232
 III.Tasks

 Running QuerySet

You can save the extracted nodes and links with specific condition as a new Workfile by hitting

Run button. Click Run button, you can see a new Workfile in Workfile Tree and see a message

box as follows. Select yes, you can see the saved new Workfile directly.

You can see the new Workfile in Workfile Tree as follows.

 QuerySet vs. Selection

Selection is similar to QuerySet in that both of them can be loaded in Filter and used for

Process. But QuerySet differs from Selection in that QuerySet is not a set of extracted nodes
and links but the condition itself for extracting nodes and links as we have covered in this

section. In contrast, Selection is the set of nodes and links selected on the network map. Full
explanations are covered in “Concept of QuerySet and Selection” in “Visual Exploration of

Network – Using Selected Nodes and Links” section.

233
Using NetMiner

11. Available Actions in “Current Workfile”

Current Workfile area is for managing DataSets and ProcessLog in a Current Workfile. In this

topic we will cover DataSet Management. Using QuerySet and Selection is covered in “Process

Handling – Handling Filter”. Using Sessions is covered in “Process Handling – Managing

Sessions”.

At Current Workfile area, the conceptual data structures are intuitively displayed. There are the

four data types in a DataSet: Main Nodeset, Sub Nodesets, 1-mode Network Data and 2-mode

Network Data. 1-mode Network data items are expanded from the Main Nodeset and 2-mode

Network data items are also expanded from each Sub Nodeset. You can see how many nodes

and attributes are in a Nodeset by checking the numbers in the appended parenthesis for

Nodeset data items. Full description about the work environment of Current Workfile is covered

in “Concepts-Work Environment”.

• Opening Data Editing Session

You can open and edit each data item in Current Workfile by selecting “Edit” in mouse right

click menu or just by double clicking the item. You can see the Data Editing Session in the right

side of the panel.

• Creating Data Item

You can create new data item using main menu “Data >> Create New Item”. But you can also

create new data item using mouse right click menu. First, select the type of data item to create

using mouse left click and select “New” in the mouse right click menu. For example, if you are

to create 1-mode Network data item, first select one of the 1-mode Network data items in

Current Workfile then select “New” in the mouse right click menu. Main Nodeset data item
cannot be created in a Workfile. A Workfile can include only one Main Nodeset data. So, there

234
 III.Tasks

is no “New” menu in the mouse right click menu for Main Nodeset data item.

• Importing / Exporting Data

You can import/export data using main menu “File >> Import/Export” or “Data >> Import Data

Item / Export Data Item”. But you can also import/export data using mouse right click menu in

Current Workfile. First, select the type of data to import and select “Import” in the mouse right

click menu. For example, if you are to import 1-mode Network data, first select one of the 1-

mode Network data and then select “Import” in the mouse right click menu. You can import

source data according to the format of data file. Full descriptions about how to use the dialogue

box for import/export data is covered in “Data Handling – Importing Data File /Exporting

Data”.

• Miscellaneous Functions

You can change the name of each data item. Select the data item to change the name and then

use the mouse right click menu “Rename”. You can change it just by hitting F2 key.

You can delete data item. Select the data item to delete and then use the mouse right click menu

“Delete” or hit “DEL” key. You can select multiple Workfile(s) using mouse left click with

CRTL, SHIFT key.

You can check the information of the properties of the selected data item and add memo to it.

Select data item and then select “Properties” in the mouse right click menu. You can see a new

dialogue box as follows.

235
Using NetMiner

You can change the order of data items in 1-mode Network data or 2-mode Network data by

using mouse drag and drop. Select the data item to move and drag it to the new position and

drop it.

236
 III.Tasks

12. Available Actions in “Workfile Tree”

Workfile Tree represents the structure of Workfiles in a Project by Tree Structure. You can see

the analysis history in Workfile Tree and manage the Workfiles intuitively. Transforming data or

importing new data makes a new expansion in the Tree Structure.

• Creating New Workfile

You can create a new Workfile using main menu “File >> New >> Workfile”. But you can also

create a new Workfile using mouse right click menu in Workfile Tree. Select “New” in the

mouse right click menu in Workfile Tree.

• Opening/Closing Selected Workfile as Current Workfile

You can open a selected Workfile as a Current Workfile by mouse right click menu “Open”.

You can do this just by double-click or hitting “Enter” key. “Close” in mouse right click menu

does not delete the Workfile but literally “close” the Current Workfile. The Current Workfile

will be empty if you select “Close”.

• Duplicating Workfile

You can duplicate the selected Workfile in the same level in the Tree Structure. Select a

Workfile to duplicate and then select “Duplicate” in the mouse right click menu.

• Importing / Exporting Workfile(s)

You can import (or export) Workfile(s) conveniently using mouse right button menu. These

functions can be done in main menu “File >> Import >> Workfile(s)” or “File >> Export >>
Workfile(s)”. Full descriptions are covered in “Tasks – Data Handling – Handling Workfile”.

237
Using NetMiner

• Changing Level in Tree Structure Using Mouse

You can change the level of selected Workfile in the Tree Structure by mouse Drag-and-Drop.

If you want to make a selected Workfile expanded as ‘child’ from a selected Workfile in the

Tree Structure, drag the Workfile to be (‘child’) and drop it under the Workfile to be parent in

the Tree Structure as shown below.

If you want to make a selected Workfile to be the same level with other Workfile, drag the

selected Workfile and drop it under the Workfile name to be the same level as follows.

• Miscellaneous Functions

You can change the name of a selected Workfile using mouse right click menu “Rename” or

just by hitting F2 key.

You can delete selected Workfile(s) using mouse right click menu “Delete” or just by hitting

“DEL” key. You can select multiple Workfile(s) using mouse left click with CRTL, SHIFT key.

The level of the child Workfile(s) of the deleted Workfile will be changed as the level of the

deleted Workfile.

238
 III.Tasks

You can check the information of the properties of the selected Workfile and add memo to it.
Select a Workfile and then select “Properties” in the mouse right click menu.

239
Using NetMiner

240
 III.Tasks

III. Tasks
Part 2. Session Based Process Handling

1. Selecting Module and Creating Session

2. Handling Process Panel Options for Running Process

3. Generating and Browsing Outputs in a Session

4. Managing Sessions

5. Handling Filter

6. Keeping Data Integrity in ProcessLog

7. Process Handling Examples

241
Using NetMiner

1. Selecting Module and Creating Session

A Session is created when a Module is selected in main menu. Let’s create a Session for
example. We will use “02.Org_Net_Tiny2.nmf” for the examples in this section.

Create a Session by selecting main menu “Analyze >> Neighbor >> Degree” as follows. We

selected Degree Module.

You can see the created “Degree” Session as follows.

242
 III.Tasks

A new item named “Degree” is expanded from Sessions in Current Workfile in the red-lined

box in the above picture.

You can create multiple Sessions for a Module. We will create three Sessions for Degree

Module. Select main menu “Analyze >> Neighbor >> Degree” repeatedly three times. You

can see the Sessions as follows. Serial number is attached to the name of each Session. You

can shift from a Session to other Session by selecting the upper tab as follows. You can check

the created three Sessions in Current Workfile.

243
Using NetMiner

2. Handling Process Panel Options for Running Proc

ess

Control Panel which is for controlling options for outputs is composed of four panels: Process,

Display, Select and Inspect. You can move among these panels by selecting inner tab. Process

Panel contains options which must be selected before running process. But other three Control

Panels contain options which are for the generated visualization outputs. In this topic, we are

going to cover Process Panel options.

Process Panel is composed of six Control Items: Filtering, Input, Pre-process, Main Process,

Post-process and Output. The five Control Items (Input, Pre-process, Main Process, Post-

process and Output) are the standardized common Control Items for processing every Module

in NetMiner. Exceptionally, in Visualize Module which is only for visualization, only

Filtering Control item is contained in Process Panel. Input Control Item is contained in

Display Panel for Visualize Module

All the five standardized common Control Items are not required for all the Modules. Only

the required Control items are activated in Process panel.

For example, we will practice using “Analyze >> Position >> Blockmodel (conventional)”

Module which requires all the five standardized common Control items.

First, create a Blockmodel Session as follows.

244
 III.Tasks

In the Control items in Process panel, it is required to specify options for fully-expanded

Control items and manipulation of options in the collapsed Control items is optional. As you

see, Input and Main Process Control items are required. Collapse all the Control items as

follows. You will find that the six Control items are quite-intuitively design for users.

The red lined five Control items are the standardized common Control items.

In this section, we will cover these five standardized Control items. Full description about
Filtering Control item is covered in “Handling – Filter”.

 Handling Input Control Item

Input Control item is for selecting data in Current Workfile for running process. The required
data type to be selected is different for each module. In one Module, you can select only one 1-
mode Network data and in another Module, you need to select one 1-mode Network data and

245
Using NetMiner

one attribute data at the same time. In some Modules, you can select only one data item,

whereas in other Modules, you are allowed to select multiple data items. The required input

data for each Module is described in “Concept – Data Structure – Data Item”.

We will use the Blockmodel Session for the example in this topic which we used in the previous

topic. The Blockmodel Session requires 2 data items as input data: 1-mode Network data (red

lined box) and an attribute data (blue lined box) as follows.

 Handling Pre-process Control Item

Pre-process Control item contains pre-required options before the main processing in a Session.

Dichotomization option for transforming weighted network data to unweighted network data or
Symmetrization option for transforming directed network data to undirected network data can

be examples. Selecting Pre-process Control item option can be required or optional according

to the selected Module.

In our example Blockmodel Session, Dichotomization is optional. You can select it or not.

246
 III.Tasks

In contrast to this case, Betweenness-Community Session (Analyze >> Cohesion >>

Community >> Betweenness) requires Dichotomization Pre-process option as prerequisite for

running process. So you can only select the criteria for dichotomization.

 Handling Main-process Control Item

Main Process Control item is for selecting core option for running process. Full descriptions

for each Module are described in “Module Reference”.

For example, in Blockmodel Session, you can select whether or not to perform Goodness of fit

Testing and choose a Goodness of Fit Index. Additionally, you can select number of iterations

and the criteria for Image Matrix and select whether or not to dichotomize for Image Matrix.

 Handling Post-process Control Item

247
Using NetMiner

Post-process Control item contains additional processing options for a selected Module. For

example, in Blockmodel Session, you can select the threshold for Role Typology of each Block

as follows.

 Handling Output Control Item

Output Control item is common to all the Modules in NetMiner. Here users get to choose

options for how outputs to be generated. Basically, NetMiner generates multiple outputs at the

same time which are required for analyzing process in the selected Module. If you exclude

some outputs which consumes a large amount of time to generate (for example, network map),

outputs can be acquired more quickly. For example, Blockmodel Session in default generates

eight outputs, but you can select what outputs you need from this Control item.

You can also select the output format in Report Format area. Internal Tab is for generating
outputs in Output Panel in NetMiner. External File is for saving the outputs as files: Output

Tables as CSV file and Main Report as HTML file. You can input the prefix for CSV files. In

Blockmodel Session, you can see the following external files if you select External File in this

option.

248
III.Tasks

249
Using NetMiner

3. Generating and Browsing Outputs in a Ses

sion

To generate outputs for a selected Module, select options in Process panel and click Run

Process button. For example, create Degree Centrality Session by selecting main menu

“Analyze >> Centrality >> Degree” as follows.

In a created Session, select 1-mode Network data “Trust” and click Run Process button at the

bottom of the Process panel.

250
 III.Tasks

Full descriptions about the options in Process panel are covered in “Handling Process Panel

Options for Running Process”. You can also refer to the Module Reference for the full

explanation about the Degree Centrality Module.

You can see that the following outputs are generated: [R] Main, [T] Degree Centrality Vector,

[M] Spring, [M] Concentric. NetMiner generates four types of output: Report, Output Table,
Network Map and Chart. [R], [T], [M], and [C], which are prefixed to the name of each
output, represents the type of output. You can move from one output to another output by

selecting the inner tab at the Output panel. [R] Main is a default output for all the analysis

module. Full explanation for each output is described in Module Reference.

251
Using NetMiner

If the type of output is table, the result can be opened in excel directly. Users are able to check

the result in excel just by selecting 'Open in Excel' function in the mouse right-click menu of
table type result.

252
III.Tasks

253
Using NetMiner

4. Managing Sessions

You can see the created Sessions in Current Workfile area as follows. You can activate a

Session using mouse double-click. The icon style is different by each type of Module:

Transform, Analyze, Statistics, Visualize and Chart.

Even if you closed the activated Session, you can see the created Sessions in Current Workfile

and activate again by mouse double-click as follows.

You can change the name of selected Session. Select a Session and use mouse right click

menu or F2 key. You can check information about the selected Session by selecting Properties

in mouse right click menu. You can also write memo for the selected Session in Memo tab. It

will be useful to record the analysis process or issues about the selected Session.

254
III.Tasks

255
Using NetMiner

5. Handling Filter

 Overview

Filter Control item is for selecting nodes and links with specific condition for running process.

The specific conditions are saved as QuerySet or Selection. In Filter Control item, you can

select QuerySet or Selection to apply.

Screenshot of Filter Control item is as follows. It is on the top of the Process panel. You need to

expand it to use it.

Filter text box represents the name of applied QuerySet or Selection. The Tree Structure

represents the structure of DataSet in Current Workfile. The attached numbers represent how

many nodes and links will be applied to run process as follows: [extracted nodes (links) / total

nodes (links)]

You can load QuerySet or Selection by hitting Load button. You can select ‘Filter’ to apply in

the dialogue box listing saved QuerySet and Selection. You can exclude the ‘Filter’ by hitting

Reset button.

You can refer to “Data Handling – Querying Data” for creating QuerySet and “Visual

256
 III.Tasks

Exploration of Network – Selecting Nodes and Links on Network Map, Using Selected Nodes

and Links” for creating Selection.

We have saved QuerySet and Selection for example as follows.

 Loading QuerySet and Selection

Create Session for this topic by selecting main menu “Analyze >> Neighbor >> Degree”.

Expand Filter Control item and hit the Load button. You can see the dialogue box for selecting

QuerySet and Selection as follows.

Select one of them to apply as ‘Filter’ and click OK.

If the selected Filter (QuerySet or Selection) is loaded, you can check it by the numbers

attached to each data item. You can see how many nodes and links will be applied for running

process in the red lined box as follows in the Filter Control item.

257
Using NetMiner

You can load QuerySet and Selection to Filter directly in Current Workfile using mouse right

click menu (“Apply”).

Create Degree Session by selecting main menu “Analyze >> Neighbor >> Degree” and select

QuerySet or Selecting to apply to Filter Control item using mouse left lick and select “Apply”

in the mouse right click menu. Note that the “Load” menu in the mouse right click menu in

Current Workfile is for launching Query Session for editing the QuerySet.

The numbers attached to each data item represents how many nodes and links will be extracted
for running process.

258
 III.Tasks

 “Run Process” with Filter

You can “Run Process” with your various selected Filters. Filter helps you to change the nodes

and links for running process very conveniently. You don’t need to make and save many

Workfiles composed of nodes and links with specific condition for analysis.

In this topic, we will use saved QuerySet for example: Finance Team and HR Team. We will

use these QuerySet in Degree Centrality Session.

Create Session by selecting “Analyze >> Centrality >> Degree”. Select QuerySet “Team =

Finance” ["Team" == "Finance"] in Current Workfile and select “Apply” in the mouse right

click menu. You can check the applied Filter as follows. You can see that 11 employees are

extracted from total 39 employees as follows.

Select 1-mode Network data “Trust” and hit the Run Process button. You can see Degree

Centrality outputs for 11 employees. If you select Concentric Map, you can see a concentric

map as follows.

259
Using NetMiner

Now we are going to perform same analysis process for HR team. Select QuerySet “Team =

HR” ["Team" == "HR"] in Current Workfile and select “Apply” in the mouse right click menu.

You can check the applied Filter as follows. You can see that 7 employees are extracted from

total 39 employees as follows. Select 1-mode Network data “Trust” and click Run Process

button.

You can see Degree Centality outputs for 7 employees in HR team. If you select Concentric

Map, you can see a concentric map as follows.

260
III.Tasks

261
Using NetMiner

262
 III.Tasks

6. Keeping Data Integrity in ProcessLog

ProcessLogs are connected to the DataSet in a Workfile. So, if the DataSet in the Workfile

changes, the integrity of the connection between the data in a DataSet and the ProcessLog that

uses the connected data will be broken. In this case, NetMiner detects and shows you which

ProcessLog is concerned to the changed data then you can select among the two options: First,

you can allow data changes and discard ProcessLog for keeping the integrity (Discard).

Second, you can expand a new Workfile with the changed data leaving the previous Workfile

as parent (Switch to Child). You can see the following message box for this selection.

263
Using NetMiner

7. Process Handling Examples

 Data Transforming : Transform >> Value >> Recode

“Recode” Session is for recoding the value of the selected data in Input Control item to a

designated new value.

Step 1: Select main menu “Transform >> Value >> Recode” for creating Recode Session and
select input data. All the data types in NetMiner can be selected for this Session. Multiple

selections are allowed. We will select 1-mode Network data “Trust”. Set the recoding

conditions as follows: 1~2  1 and 3~4  2, 5 3.

Step 2: Set Main Process Control item options. Diagonal Handling Option is for selecting
whether or not to include diagonal values in recoding process. You can give recoding condition
in Recoding Rules. Click Add button for launching dialogue box for setting recoding rule. The

values between the two numbers in the red box will be recoded to the number in blue box.

Set the first condition: “1~2  1” and click OK. You can see that the first condition is added to
the Control item as follows.

264
 III.Tasks

Continue to add the recode condition. If you finished adding all the recode conditions, click

Run Process button.

Step3: The recoded 1-mode Network data “Trust” will be generated as an output of the Session.
And you can see a dialogue box for selecting data for adding to Current Workfile. In this

example, we have generated one recoded 1-mode Network data. It will be added as a new 1-

mode Network data to Current Workfile. Click OK. If you selected multiple data in Input

Control item, you can see also multiple items in this dialogue box. Then you can select data

item which you want to add to Current Workfile.

If the transformed 1-mode Network data is symmetric, you can add it to the Current Workfile as

a 1-mode Network data item with “Undirected” attribute. You can select whether or not add it

as “directed” or “undirected” using the following dialogue box.

265
Using NetMiner

If you select “undirected”, the generated and transformed 1-mode Network data corresponds to

a white-colored upper-half of matrix type as follows. The gray area is not editable.

Step 4: You can see a new 1-mode Network data “Recoded (Trust)” in Current Workfile.

 Generating Analysis Report : Analyze>>Centrality>>

Degree

266
 III.Tasks

Degree Module is for calculating degree centrality of each node. In-Degree Centrality and Out-

Degree Centrality vectors are generated at the same time.

Step1: Create Degree Centrality Session by selecting main menu “Analyze >> Centrality >>
Degree” and select 1-mode Network data “Trust”. You can select option in Main Control item.

Click Run Process button. Full descriptions about the options in this Session are covered in

Module Reference.

Step2: You can see that four outputs are generated as follows: [R] Main Report, [T] Degree
Centrality Vector, [M] Spring-Map and [M] Concentric Map. In Main Report, you can check

the process information and output summary. Information represents the Module and the data

for this Session. Output Summary shows the mean, Standard Deviation, Maximal Value and

Minimal Value of the Degree Centrality vector. You can also check the Degree Centralization

Index. Centralization Index represents how much the high centrality values are concentrated on

small number of nodes. If the Centralization Index closes to 1, high degree centrality

concentrated on small number of nodes. [T] Degree Centrality Vector shows In-Degree

Centrality and Out-Degree Centrality of each node. You can see the Network Map of the

selected 1-mode Network. The size of each node is the Degree Centrality value. You can also

see the Centrality vector in Concentric Map. Concentric Map shows the output result intuitively.

You can inspect the Network Map generated in Analyze menu by Inspect Control panel. Process

for generating Analysis Results in Analyze menu is easier than the transformation process.
You can inspect the Network Map generated in Analyze menu by Inspect Control panel. Process

for generating Analysis Results in Analyze menu is easier than the transformation process.

 Generating Analysis Report: Analyze>>Cohesion>>Component

Component Module shows how many components are in a selected network and which

component each node belongs to.

Step1: Create Component Session by selecting main menu “Analyze – Cohesion – Component”

267
Using NetMiner

and select 1-mode Network data “Friendship”.

Step 2: You can select Minimum Size of Component and the Component Type in Main Process
Control item. Minimum Size of Component is the minimum number of nodes in a Component to

recognize it as a Component. For example, if you select the size = 2, isolate nodes are not

recognized as Component. You can select Weak Component or Strong Component as

Component Type. Weak Component does not consider the direction of connection among the

nodes in a Component. But to be a Strong Component, all the nodes in a Component must be

connected reciprocally. We will select 1 as Minimum Size of Component and “Weak

Component” as the Component Type. Click Run Process button

Step 3: The generated outputs are [R] Main, [T] Component Partition Vector and [M]
Clustered. In [R] Main, you can see the number of Component and the members of Component.

You can also check the Size and Density of each Component. In [T] Component Partition

Vector, you can see which Component each node belongs to. In [M] Clustered, you can see a

clustered network map.

 Using Generated Outputs for Another Process : in Statistics

You can use outputs generated as an input data for another Process. This topic covers the

Recursive data usage in NetMiner.

You can see “Duration” in the attribute data of Main Nodeset. It represents how many years

each employee has employed by this organization. We will now analyze the correlation between

the “Duration” and the In-Degree Centrality of each five 1-mode Network data and to find

which 1-mode Network data has the highest correlation with “Duration” in In-Degree

Centrality.

268
 III.Tasks

Step1: Create Degree Centrality Session by selecting main menu “Analyze – Degree-
Centrality” and select 1-mode Network data “Trust”. Click Run Process button.

Step2: Select [T] Degree Centrality Vector, you can see In-Degree Centrality Vector and Out-
Degree Centrality Vector. We are going to save the In-Degree Centrality Vector as an attribute

data of Main Nodeset. Select main menu “Data – Add Output Table – Main Node Attribute” as

follows.

You can see a dialogue box for selecting item to add to attribute data of Main Nodeset. Select

In-Degree Centrality and change the name as “Trust_IND”. Click OK button.

269
Using NetMiner

You can see the saved In-Degree Centrality (“Trust_InD”) in the attribute data of Main Nodeset

as follows.

Step3: In the same way, you can add the In-Degree Centrality of each five 1-mode Network
data. You don’t need to create a new Session. You can generate new outputs just by changing

the 1-mode Network data in Input Control item and click Run Process button. You need to add
five In-Degree Centrality of each 1-mode Network data as follows.

270
 III.Tasks

Step4: Now we can calculate the correlation table among the five In-Degree Centrality and
“Duration” using Statistics Module. Create Vector Correlation Session by selecting main menu

“Statistics – Correlation – Vector”. Select six Node Attribute data in Input Control item as

follows. In Main Process Control item, you can select the method to measure the correlation.

Select “Correlation” and “Pearson’s Correlation”. You can refer to Module Reference for

details in this Module. Click Run Process button.

271
Using NetMiner

Step5: You can see the correlation table in Main Report as follows.

You can check the Pearson’s Correlation values between the “Duration” and the In-Degree

Centrality of each five 1-mode Network data. You can see the significant correlation between

the In-Degree Centrality of “Trust”, “Knowledge” and “Influence” network and “Duration”.

In this way, the generated outputs in one Module are re-used as an input data for another

Process.

 Using Generated Outputs for Another Process : in Visualize

You can use generated outputs for another Process Module. In this topic, we will use the

generated output for visualization. For example, you can give style to each node using the In-

Degree Centrality vector generated in Degree Centrality Module.

We will use the sample data that was used in previous topic “Using Generated Outputs for
Another Process: in Statistics”. We have added five In-Degree Centrality vector to attribute data

of Main Nodeset.

Step1: Create Spring 2D Session by selecting main menu “Visualize >> Spring >> 2D” and
select 1-mode Network data “Trust”. Leave other options set to default and click Run Layout

button. You can see a network map.

Step3: We are going to represent the size of node on the network map in proportion to the In-
Degree Centrality value of each node. Launch the dialogue box for Node and Link Attribute

272
 III.Tasks

Styling as follows. You can select an icon in the toolbar as follows.

You can see a dialogue box as follows.

Select Node Attribute Styling tab. Then, in Size tab, select “Node Attribute” in the second pull-

down menu, “Trust_InD” in the third pull-down menu, and click Apply button.

273
Using NetMiner

 Generating Charts

You can use various charts for representing data. Process Control panel in Chart Module also

contains common Control items as Input or Main Process.

Step 1. Create Matrix Diagram Session by selecting main menu “Chart >> Matrix Diagram”.

Step 2. Select “Trust” as input data and select “Team” as Permutation Vector. We are going to
make a chart representing “Trust” network clustered by “Team”.

Step 3. In Main Process Control item, you can select various options. Select Node Label,
Border, Border Name and Colored Scale. And do not select Grid. Click Run Process button.

274
 III.Tasks

 Generating Network Map

You may know that the Sessions created by selecting Visualize Module are different from other

Sessions. In other Sessions, six Control items are commonly contained in the Process Control

panel. But in Visualize Sessions, five standard common Control items in Process panel does not

exist. Display Control panel is default Control panel for Visualize Sessions. In Display Control

panel, selecting 1-mode Network data and layout algorithm is required. You can draw network

map by clicking Run Layout button. Navigation Control item is for controlling network map

panel as Zoom In (Out), Navigation etc. Node Style and Link Style Control item is for giving

style to nodes and links on the network map.

You can analyze network data in an integrated environment for visual exploration. Full

descriptions about Network Map Visualization are covered in another chapter “Visual

Exploration”.

 Generating Mining Report: Mining >> Classification

>> SVMs

Node attributes can be used as input data for Mining and this process is expatiated with the

following example.

Suppose there are a main nodeset and its node attributes, which are either numerical or text

attributes. When numerical attributes are relevant to the values of a text attribute, a user can use

the numerical attributes to predict missing values of a text attribute.

For instance, numerical attributes named ‘age’ and ‘duration’ of an employee are likely to be

related to his or her ‘position,’ which is a text attribute. As such, classification algorithms in

Mining can be utilized to predict the missing values of a ‘position’ attribute by using the

numerical attributes (i.e. ‘age’ and ‘duration’).

275
Using NetMiner

276
 III.Tasks

III. Tasks
Part 3. Visual Exploration

1. Overview

2. Drawing Network Map

3. Controlling Views

4. Selecting Nodes and Links

5. Styling Nodes and Links

6. Using Selected Nodes and Links

7. Inspecting Analysis Results in Network Map

8. Recording Visual Exploration

277
Using NetMiner

1. Overview

Network Map in NetMiner was designed to support user’s visual exploration of network data

beyond network plain data representation. You can draw network map using various layout

algorithm and give various style to each node and link. You can do the following tasks using

network map in NetMiner.

- Drawing Network Map

Drawing Network Map is giving optimal coordinates to each node of network map. You

need to select a 1-mode Network data and a layout algorithm for drawing it. You can select

specific option for each layout algorithm.

- Controlling Views

Basic features for controlling network map are implemented: Zoom In/Out, Flipping and

Panning, Rotating and so on.

- Selecting Nodes and Links

You can select nodes and links in various ways. You can select nodes and links simply

using mouse drag or using specific condition.

- Styling Nodes and Links

You can assign various styles to each node and link, and change shape, color and size of

each node and arrow shape, color and width of each link. Also, you can also give each node
and link an optimal style automatically by attribute.

278
 III.Tasks

- Using Selected Nodes and Links

You can use selected nodes and links in various ways for analysis and visualization.

- Inspecting Analysis Results in Network Map

You can inspect the output results in network map by applying the output results to network

map interactively. It gives you an intuitive understanding of the analysis results.

- Recording Visual Exploration

NetMiner supports recording and editing the network map exploration. This feature can be

applied to not only the exploration but also the presentation.

279
Using NetMiner

2. Drawing Network Map

Drawing Network Map assigns optimal coordinates on each node of network map. You need to

select a 1-mode Network data and a layout algorithm for drawing it, and select specific option

for each layout algorithm.

 Node Layout Algorithms

NetMiner is equipped with various node layout algorithms. You can select one of these

algorithms that fit your need of data analysis and representation. Those are classified as 2D

network map and 3D network map as follows.

• 2D Network Map

- Spring
 Kamada & Kawai

 Stress Majorization

 Eades

 Fruchterman & Reingold

 GEM
 HDE

- MDS
 Classical MDS

 Non-metric MDS

 Kn-MDS

- Clustered
 Clustered Eades

280
 III.Tasks

 Clustered Co-La

- Layered
 Dig-Cola

- Circular
 Circumference

 Concentric

 Radial

- Simple
 Fixed

 Random

• 3D Network Map

- Spring
 Kamada & Kawai

 Eades

- MDS
 Classical MDS

 Non-metric MDS

- Clustered
 Clustered Eades

Full description about each layout algorithm is covered in Module Reference. You need to refer

to this book for selecting specific options in each layout algorithm. We give you screenshot of
each layout algorithm as follows.

281
Using NetMiner

• 2D Network Map

- Spring

 Kamada & Kawai  Stress Majorization

 Eades  Fruchterman & Reingold

 GEM  HDE

282
 III.Tasks

- MDS
 Classical MDS (without links)  Non-metric MDS (without links)

 Kn-MDS (without links)

２８３
- Clustered
 Clustered Eades  Clustered-CoLa

- Layered
 Dig-Cola
 III.Tasks

- Circular
 Circumference  Concentric (without links)

 Radial

- Simple
 Fixed  Random

２８５
• 3D Network Map
- Spring
 Kamada & Kawai  Eades

- MDS
 Classical MDS  Non-metric MDS
 III.Task

Clustered
 Clustered Eades

287
Using NetMiner

 Link Layout Algorithms

NetMiner 4 contains link layout algorithms that are designed to further improve the visibility

and succinctness of graph visualizations along with node layout algorithms.

NetMiner 4 currently contains the following link layout algorithms:

 Basic

 Divided Edge Bundling

• Divided Edge Bundling

 Methods for Drawing Network Map

There are two ways for drawing network map in NetMiner. One is using Visualize Module

directly and the other is using Analyze Module. Most Analyze Modules generate network map

as one of the outputs.

288
 III.Task

• Drawing Network Map Using Visualize Menu

Step1: Select main menu “Visualize”. You can see categorized various network map layouts
as follows. Select one of those which fit your analysis needs. For example, select main menu

“Visualize >> Spring >> 2D”.

Step2: You can see Spring 2D Session is created. There are thirty nine nodes distributed
across the Output panel randomly, with various options for generating network map in the

Display Control panel on the right. Modules categorized as Visualize do not have five

standardized common Control items in Process Control panel.

Step 3: Select 1-mode Network data for visualization and select various options for the
selected layout algorithm and hit the Run Layout button. In this example, select 1-mode

Network data “Trust” and select “Spring-Kamada&kawai” as the layout algorithm. You can

289
Using NetMiner

see the following network map in the Output panel.

Methods about selecting layout algorithm and options for drawing network map are covered

in “Handling Display Control Panel for Drawing Network Map”.

If you have ‘Huge’ size (1,000,000 nodes) Network data which requres much time complexity

to draw network map, you can draw network map with separated two steps using Layout and

Drawing module in Visualize menu as follows.

Step1: Select main menu “Visualize>>Layout” as follows. “Layout” module generate

coordinates of the nodes in the network map.

You can select input data, layout algorithm and options as you can see in other ‘Visualize’

modules as follows.

290
 III.Task

Select “Trust” as input data and leave other options as default. Click Run Process. You can see

the output of this module. If you select Node Position inner tab as follows, you can see the

coordinates of the nodes. You can add these coordinates vectors as node attribute data for

using in drawing network by mouse-right click menu “Add to Workfile >> Node Attribute”.

Check X position and Y Position in the ‘Select Attribute’ dialogue box as follows.

291
Using NetMiner

Now you can check the added attribute in Node Attribute data item as follows.

Step2: Select main menu “Visualize>>Drawing” as follows. In “Drawing” module, you can
select input data and X, Y Position for drawing network map.

292
 III.Task

You can select the added attribute data of node position generated in Step 1 as follows.

Click Run Process for drawing network map.

If you want to give style to nodes and links, you can use Display or Select panel. Details are

coverd in “Styling Nodes and Llinks” section.

293
Using NetMiner

• Drawing Network Map Using Analyze Menu

Step1:’Analyze’ category in main menu contains various general modules. Most Modules
contain network map as one of each Module’s various outputs. So users can get the network

map by running Modules categorized as Analyze. Network map generated in Analyze Modules

can be explored in Inspect Control panel. You can apply the analysis result to the network

map for intuitive analysis in Inspect Control panel. Select main menu “Analyze >>

Connection >> Shortest Path” for creating Shortest Path Session for example.

Step 2: Select 1-mode Network data “Trust” and click Run Process button. You can see three
outputs “[R]Main, [T] Distance Matrix, [M] Spring” are generated in Output panel. Select

[M] Spring, you can see the network map as follows. You can change the layout algorithm or

the styles of nodes and links in Display Control panel.

Step 3: Select Inspect Control panel. You can see Control items for applying the analysis
result to network map at the right side. Select “CY001” as Source Node, “CY013” as Target

Node. And click Submit button. The value “2” means the Distance Matrix value between the

294
 III.Task

selected two nodes. You can see the two nodes and the shortest path between these two nodes

on the network map. In this way, you can apply the analysis results on network map and can

understand the analysis results intuitively. You can also change the layout algorithm and the

style of nodes and links in Display panel. If you are to select another 1-mode Network data,

you need to move to Process Control panel. Note that Modules in Analyze menu contain Input

Control item in Process Control panel, but Modules in Visualize menu in Display Control

panel.

 Saving Coordinates

You can save coordinates of nodes and draw network map using the saved coordinates. It will

be especially useful if you want to change the coordinates of nodes using mouse drag for some

reasons. If you draw network map using layout algorithm, the changed coordinate will be lost.

But if you save them and draw using the saved coordinate, you can get the changed network

map again. Let’s do this work for example.

Step 1: We will use “Trust” Network map. We are going to drag the “CY084” to other position
using mouse drag as follows.

Step 2: Now save the coordinate by selecting “Save Coordinates” in the mouse right click
menu on the network map.

295
Using NetMiner

You can see the following message box.

You can see the saved coordinates in the attribute data of Main Nodeset as follows.

You can save the coordinates using the following toolbar easily.

Step 3: Now we can draw network map using the saved coordinates. Select main menu
“Visualize >> Simple >> 2D” and select “Fixed” as Layout Algorithm. Select 1-mode Network

data “Trust”. Select “User Defined” as the Initial Coordinate. The dialogue box for selecting

coordinates will pop-up. You can select the saved attribute variable in each pull-down menu.

296
 III.Task

Click Run Layout button for drawing network map based on the selected saved coordinates.

297
Using NetMiner

3. Controlling Views

 Overview

You can get more information about network map by applying many kinds of view: zoom

in/out, rotating, changing focus, panning and so on. Because the difference of dimension cause

the difference of manipulating, This section is divided in two parts : 2D and 3D view.

 2D view

• Navigation
Navigation Control item : Zoom in the network map, and you can drag the red colored
square in the Navigation Control item to navigate the network map.

• Zoom (in/out)
Mouse : When the mouse pointer is focused on the network map, you can use mouse
wheel to zoom in(or out).

Short-key : You can use keyboard. “+” and “-“ key in the number pad with CTRL key
will work.

298
 III.Task

Navigation Control item : You can zoom in (or out) using the slide bar in the Navigation
Control item. You can also input the zooming ratio in the text box. If you click ‘Fit to

Screen’ button, you can back to the default size. If you are to enlarge the size of node and

label only, you can use the arrow buttons in the Navigation Control item as follows.

• Changing size of map objects simultaneously (nodes, links, and

labels)
Mouse : When the mouse pointer is focused on the network map, you can change size of
nodes, links, and labels simultaneously by mouse wheel with CTRL key

Navigation Control item : You can do same operation mentioned above by using the
arrow buttons in the Navigation Control item.

• Panning
Mouse : Zoom in the network map, and you can use panning mode by dragging with
mouse click.

• Flipping
Main Menu : You can flip the network map vertically or horizontally. Select main menu
“Map >> Flip >> Flip Horizontal” or “Map >> Filp >> Flip Vertical”.

 3D view

• Changing the Focus Node

Mouse : If you double-click a node, the map focus is moved to the corresponding node.

• Zoom (in/out)
Mouse : When the mouse pointer is focused on the network map, you can use mouse

299
Using NetMiner

wheel to zoom in(or out).

Short-key : You can use keyboard. “+” and “-“ key in the number pad with CTRL key
will work.

• Changing size of map objects simultaneously (nodes, links, and

labels)
Mouse : When the mouse pointer is focused on the network map, you can change size of
nodes, links, and labels simultaneously by mouse wheel with CTRL key

• Rotating
Mouse : Drag with mouse right-click.

• Panning
Mouse : Drag with mouse left-click.

300
 III.Task

4. Selecting Nodes and Links

In exploratory analysis process on the network map, you may frequently need to select nodes

and links for styling or another analysis using the selected nodes and links. Knowing various

ways to select nodes and links on network map will help you to work efficiently on the network

map in NetMiner.

Various ways to select nodes and links in NetMiner can be classified into two categories: Visual

Selection and Non-Visual Selection, or more simply speaking, mouse-drag and menu selection,

respectively. You can also select nodes with specific condition. The color of selected nodes and

links will be changed to yellow.

 Visual Selection

 Selecting & Deselecting Nodes and Links Using Mouse

You can select nodes and links on network map using mouse. You can select each node and link

by mouse left-click. If you select a link, two nodes at the end of the link will be selected either.

Multiple nodes and links are selected by mouse drag with Shift key. You can select additional

nodes and links with CTRL key. ALT key can be used for excluding the selected nodes and

links.

301
Using NetMiner

 Selecting Each Node Using Select Control Panel

If there are too many nodes on the network map to select specific nodes just by mouse click,

you can select nodes in Selection Control item using Select Panel by choosing each node on the

nodelist. The Selection Control item is for representing the selected nodes on the network map.

But you can select nodes in this Control item as above.

 Selecting Neighbors of a Selected Node

You can select neighbors of a selected node (focus node) on network map. The function will be

run if you select a node and select “Select Neighbor” in the mouse right click menu on the

focus node. When selecting neighbors of a node, direction of the link is considered. If you

select 'In Direction', the neighbors which send the links to the focus node are selected. For the

'Out Direction' option, the neighbors which receive the links from the focus node are selected,

and for the 'Both Direction' option, the whole neighbors of the focus node are selected

disregarding the direction of the links. When you repeat the process, the neighbors of 2 distance

are selected. You can select neighbors within ‘user-specified distance’ by repeating this process.

While repeating the ‘select – neighbor’ menu, NetMiner 4 asks which direction of neighbors
you want to select, again. The last chosen direction is applied to the selection.

302
 III.Task

You can see that the neighbors of the focus node are selected.

 Deselecting Isolate Nodes and Pendants

If isolate nodes or pendants are selected with other nodes, you can exclude them from the

selection conveniently. Select “Deselect Isolate Nodes” or “Deselect Pendants” in the mouse

right click menu when you have selected nodes including isolate nodes or pendants.

303
Using NetMiner

 Non-Visual Selection

 Selecting Nodes with Specific Condition

You can select nodes and links with complex, specific condition using Condition Control item

in Select Control panel. For example, if you are to select nodes whose gender is “male” and the

age of whom is over thirty, you can select the nodes by writing the following expression:

"Gender" == "male" and "Gender" >= 30

Variables, Operators and the available values for each variable can be selected by mouse click.

Click Run Query button to select the nodes on the network map. You can see that the color of

the selected nodes and links change to yellow as follows. You can refer to “Querying Data in

Data Handling” section for more specific descriptions about writing the expression because the

user interface of Condition Control items is similar to that of Query Session.

You can check the selected nodes in the Selection Control item. The selected nodes can be

saved as Selection and used in another Process Module. Full description about saving and using
the selected nodes and links is covered in next section “Using Selected Nodes and Links”.

304
 III.Task

 Selecting Links by Manipulating Links’ Threshold

value

You can show or hide links according to the weight (or their other numerical attribute value) of

each link on the network map with various threshold options using Link Threshold Control item

in Select Control panel. The links shown on the network map have the weight (or the numerical

attribute), which is between the minimum value and the maximum value specified in the Link

Threshold Control item. The minimum value and the maximum value also can be set in the blue

slide bar.

You can select Numerical Link Attribute as @Weight or other numerical link attribute data of

each link on the network map in Numerical Link Attribute area box. When you select a link

attribute, the scale of that attribute is presented in the slide bar of 'Link Threshold' box. You can

also decide to include or exclude the links with missing values in this option.

You can run again the current Process with the selected links on the network map if you select

the Run with the showing links button. You can also save the linkset on the network map as

QuerySet using “Save Threshold as QuerySet” button. Full description about saving and using

the selected links is covered in next section “Using Selected Nodes and Links”.

305
Using NetMiner

5. Styling Nodes and Links

You can give various styles to the selected nodes and links. And you can give optimal style to

the nodes and links on the network map according to the attribute of nodes and links

automatically. This section covers seven topics as follows: Styling Selected Nodes, Styling

Selected Links, Node Attribute Styling, Link Attribute Styling, Node Labeling, Link Labeling and

Saving Styles. You can move to the topic which you need. The process of 3D map styling is

same as the process of 2D.

 Styling Selected Nodes

You can use the Node Style Control item for giving style to node(s).

If you double click the selected node(s) in the Node List, you can see a dialogue box for giving

various styles. You can select multiple nodes using CTRL, SHIFT key with mouse left click.

Details about the dialogue box will be covered in following explanation. Frequently used

method for giving style to the selected nodes is using mouse on the network map. Generally, the

Node List is used when there are too many nodes on the network map to select a specific

node(s) using mouse. You can change the label of each node using Show Node Labels. Full

descriptions will be covered in “Node Labeling”. You can give optimal style to each node

306
 III.Task

automatically using “Node Attribute Styling” dialogue box.

Select nodes using mouse on the network map. You can select multiple nodes with SHIFT key.

Check the mouse right click menu on the network map. “Node Style…” is for giving style to a

single selected node. “Multiple Nodes Style…” is for giving style to multiple selected nodes.

However, the dialogue box for giving style is same for these two menus.

- Preview: To preview the node which will be shown on the network map with the selected
styles.

307
Using NetMiner

- Shape: To select shape of node.

- Surface: To select the color of node as single color or gradation color and to represent the
node by image file.

- Size: To change the size of node.

- Border: To change the border thickness and color.

- Label: To select the position of node label and whether or not to give border to the label.
You can also select the background color and font of the label.

You can select the items independently by selecting the check box of each item.

You can now change the multiple properties of nodes such as shape, surface, size, border, and

label all together by checking the boxes of property types and selecting their attributes of your

choice.

For example, select 1-mode Network data “Trust”. Select a node “CY013” using mouse left

click and select “Select Neighbor” in the mouse right click menu as follows.

Now we are going to change the style of the selected nodes. Select “Multiple Nodes Style” in

the mouse right click menu on the network. In the dialogue box, select options as follows and

click OK: shape – square, Surface – Single blue color, Size- 15.

308
 III.Task

A map applied with new style of selected nodes is shown below.

 Styling Selected Links

You can use the Link Style Control item for giving style to link(s).

309
Using NetMiner

If you double click the selected link(s) in the Link List, you can see a dialogue box for giving

various styles. You can select multiple links using CTRL, SHIFT key with mouse left click.

Details about the dialogue box will be covered in following explanation. Frequently used

method for giving style to the selected links is by using mouse on the network map. Generally,

the Link List is used when there are too many links on the network map to select a specific

link(s) using mouse. You can change the label of each link using Show Link Labels. Full

descriptions will be covered in “Link Labeling”. You can give optimal style to each link

automatically using “Link Attribute Styling” dialogue box. If Link List is not selected

(unchecked), the links on the network map will be disappeared. You can select it using toolbar

as follows. Link width, arrow shape also can be selected in this tool bar. You can do this work

using main menu “Map >> Show” also.

Select links using mouse on the network map. You can select multiple links with SHIFT key.

Check the mouse right click menu on the network map. “Link Style…” is to set style of a single

selected link. “Multiple Links Style…” is for setting style of multiple selected links. However,

the dialogue box for giving style is same for these two menus.

310
 III.Task

- Preview: To preview the link, which will be shown on the network map with the selected
styles.

- Head: To change the shape of arrow.

- Stroke: To change the color, shape and width of the link.

- Label: To select the position of link label and whether or not to give border to the label.
You can also select the background color and font of the label.

For example, select 1-mode Network data “Trust”. Select a node “CY013” using mouse left

click and select “Select Neighbor” in the mouse right click menu as follows.

311
Using NetMiner

Now we are going to change the style of the selected links. Select “Multiple Links Style” in the

mouse right click menu on the network. In the dialogue box, select options as follows and click

OK: Stroke – color: blue, width: 3.

A map applied with new style of selected links is shown below.

312
 III.Task

 Node Attribute Styling

You can set optimal style to each node according to the attribute of node. For example, if there

is an attribute representing gender, you can give optimal style nodes on the network map

automatically as follows: male: color - blue, shape - square, female: color - red, shape - circle.

Node And Link Attribute Styling dialogue box is designed for this kind of usage. You need to

prepare the attribute data for using Node Attribute Styling dialogue box.

Click the Node Attribute Styling button in the Node Style Control item. You can launch this

dialogue box using toolbar as follows.

313
Using NetMiner

Following is the screenshot of Node And Link Attribute Styling dialogue box.

Node And Link Attribute Styling dialogue box is composed of two tab: Node Attribute Styling,

Link Attribute Styling.

Node Attribute Styling tab is composed of four tabs: Quick, Shape, Color and Size. Quick is for

changing shape, color and size simultaneously.

- Variable: List in the second pull-down menu changes according to the selection made on
the first pull-down menu. Available variable options in the first pull-down menu are as

follows.

 Node Attribute: You can give same style to the nodes with same attribute. You can select

attribute variable in the second pull-down menu.

314
 III.Task

 2-mode Network Data: You can give same style to the nodes with same weight in the 2-

mode Network data. You can select the Sub Nodeset as the criteria for giving different style.

 All Nodes: You can change the style of all the nodes at the same time. Second pull-down

menu will be disabled in this case.

 Each Node: You can give different style to each node. Second pull-down menu will be

disabled in this case.

- Setting: You can select options for the style.

 Inverse Order: For visualizing style, change of order can be useful. For example, nodes

with high weight are given biggest size as default. However, you can set smallest size to the

node with highest weight, if selecting Inverse Order option.

 User-defined Color scale: You can set the start color and the end color of the color scale.

This option is valuable when the chosen attribute is a continuous variable.

 Min/Max: The size of nodes are set between the minimum value and the maximum value

set in this option.

- Preview: You can preview the styles given to each item. You can change the style by
launching the dialogue box for giving style by mouse double click.

- Default: Recover the user-defined style to automatically generated style.

- Random: To set random style to each item.

- Copy to Clipboard: Screenshot of Preview is copied to clipboard. You can use it as legend
of the network map.

You need to click Apply button to apply the selected style to the nodes on the network map.

Only the styles in the selected Tab are applied. You need to click Apply button for each Tab.

For example, we will give style nodes on the “Trust” network map. We will give different shape

315
Using NetMiner

to the nodes with different gender, different color to the nodes with different team and different

size to the nodes with different position.

First, in the Shape tab, select “Gender” as the second variable. You can see that circle and

triangle style is given to each attribute, respectively. Double click male and change the shape to

square. Similarly, change the shape of female to circle. Click Apply.

Second, in Color tab, select “Team” as the second variable and click Apply.

Third, in Size Tab, select “Both” in Type and select “Position” as the second variable. You need

316
 III.Task

to check the Inverse Order to give optimal style. You can change the Min/Max values, which

define the range of diameter of shape of nodes. Click Apply.

A network map with new style setting will be shown below.

317
Using NetMiner

 Link Attribute Styling

You can give optimal style to each link automatically according to the weight and attribute of

link. For example, you can give different width to each link according to the weight of the link.

Click the Link Attribute Styling button in the Link Style Control item. You can launch the

dialogue box using toolbar as follows.

Following is the screenshot of Node And Link Attribute Styling dialogue box.

Link Attribute Styling tab is composed of three tabs: Width, Color and Head.

- Variable: List in the second pull-down menu changes according to the selection made on
the first pull-down menu. Available variable options in the first pull-down menu are as

follows.

 Link Attribute: You can give same style to the links with same link attribute or weight. You

can select attribute variable in the second pull-down menu.

 Multiple Link: You can give different style to the multiple links

318
 III.Task

 All Links: You can change the style of all the links at the same time. Second pull-down

menu will be disabled. You can use toolbar for this function easily as follows.

 Each Link: You can give different style to each link. You cannot select second variable.

- Setting: You can select options for the style.

 Inverse Order: For visualizing style, change of order can be useful. For example, links with

high weight are given the thickest widths as default. However, you can set thinnest width to

the link with highest weight, if selecting Inverse Order option.

 Min/Max: Activated only in Size Tab.

 User-defined Color scale: You can set the start color and the end color of the color scale.

This option is available only in the Color tab.

- Preview: You can preview the styles assigned to each item. You can change the style by
launching the dialogue box for giving style by mouse double click.

- Default: Recover the user-defined style to automatically generated style.

- Random: To set random style to each item.

- Copy to Clipboard: Screenshot of Preview is copied to clipboard. You can use it as legend
of the network map.

You need to click Apply button to apply the selected style to the links on the network map. Only

the styles in the selected Tab are applied. You need to click Apply button for each Tab.

For example, we will give style links on the “Knowledge” network map. We will give different

color to the links with different Category, different width to the links with different weight.

319
Using NetMiner

First, in the Color tab, select “Category” as the second variable and click Apply.

Second, in the Width tab, select “Weight” as the second variable and click Apply.

320
 III.Task

You can get a network map as follows.

 Directed Link Styling

You can select style about directional link. You can change it using toolbar as follows. You can

do this work using main menu "Map >> Show" also.

default style is as follows.

321
Using NetMiner

If you check this option, link style is changed as follows.

 Node Labeling

Default node label is the node ID. However, you can change the node label using other node

attributes such that each node can be represented by its other attribute value.

You can hide the node label on the network map by deselecting the Show Node Labels check

box as above. You can select Main Node attribute variable in the pull-down menu as the label

of each node on the network map. It can be done by selecting main menu “Map >> Show >>

Node Label” or using toolbar as follows.

 Link Labeling

Link label is not represented as a default setting but you can represent it as weight or attribute

322
 III.Task

of link.

Select Show Link Labels check box to represent link label. You need to select 1-mode Network

data in the first pull-down menu and the weight (or the attribute of link) in the second pull-

down menu. You can select any 1-mode Network data in the Current Workfile in the first pull-

down menu even if the 1-mode Network data was not selected as input data for drawing

network map. If you select 1-mode Network data in the first pull-down menu which was not

selected as the input data, the represented link label for each link is “Missing Value”. You can

change the link label using main menu “Map >> Show >> Link Label” or toolbar easily as

follows.

For example, we will select two 1-mode Network data “Trust” and “Knowledge” as input data

for drawing network and then give link label. First, select the Show Link Labels check box.

Select “Trust” in the first pull-down menu and select “Weight” in the second pull-down menu.

You can get the following network map representing the link label as weight of “Trust” network

map. You can see link labels representing “Missing Value”. Links representing “Missing Value”

as their label does not exist in “Trust” network, but only in “Knowledge” network. Note that

even if multiple 1-mode Networks can be represented at one time for drawing network, you can

323
Using NetMiner

represent labels of the selected just one 1-mode Network data at one time.

Let’s look closer on the network map. The link between two nodes CY171 and CY229

represents two labels. In this case, the link from CY171 to CY229 does not exist in “Trust”

network. This link exists only in “Knowledge” network. But the weight of link from CY229 to

CY171 is 5. Additionally, you can notice that the link from CY066 to CY171 does not exist both

in “Trust” network and “Knowledge” network.

Now let’s change the option for representing other attribute of link. Select “Knowledge” in the

first pull-down menu and select “Meeting Time” in the second pull-down menu. You can see

that the “Meeting Time” is represented as the label of each link as follows.

324
 III.Task

 Saving Styles

You can save the style of each node and link on the network map. You can use the saved style

in another Session and if you run process again, the saved style will be applied automatically to

the nodes and links.

Step 1: First, we will assign various styles to the nodes and links on the “Trust” network map.
We will give different shape for different gender, different color for different team and different

size for different position. After that, select “Save Styles >> Both” in the mouse right click

menu on the network map.

325
Using NetMiner

You can see the following message box.

Step 2: Now we are going to create a new session and check whether or not the saved styles are
applied to nodes and links on the network map in a new Session. Create Session by selecting

main menu “Visualize >> Clustered >> 2D”. Select 1-mode Network data “Trust” and “Team”

as Node Attribute Vector and then click Run Layout button. You can see that the saved styles are

applied to the nodes and links on the network map. In this way, you can save and reuse the

saved styles conveniently.

You can use the icon in toolbar as follows.

326
 III.Task

6. Using Selected Nodes and Links

You can use the selected nodes and links in various ways for analysis and visualization.

Previous section “Selecting Nodes and Links” is pre-requisite for this section.

 Concept of QuerySet and Selection

You can extract nodes and links with complex condition using Query Session and use them for

analysis and visualization. You can also save the selected nodes as Selection and use it in

various ways. First of all, we need to know the concept of QuerySet and Selection.

QuerySet is a condition itself which was made in Query Session created by selecting main
menu “Tools >> Query Composer”. The condition expressed by weight threshold designated in

Weight Threshold Control item in Select panel can also be saved as QuerySet. Note that

QuerySet is not the DataSet composed of nodes and links but the condition itself. You need to

apply the QuerySet to Filter to run process with the selected nodes and links expressed in the

QuerySet.

QuerySet can be saved and you can edit the QuerySet in Query Session by loading it. The saved

QuerySet can be used in the Sessions in the Current Workfile. The conditions in a QuerySet can

be made for any data item in Current Workfile: Main Nodeset, Sub Nodeste, 1-mode Network

data and 2-mode Network data. You can give condition to nodeset data item about the attribute

of nodes and to linkset data item about the weight and attribute of link. You can refer to “Data

Handling – Querying Data” for more details about making QuerySet. And you can refer to

“Using Saved QuerySet and Selection in Filter” for applying the saved QuerySet to each

Module for running process.

Selection, in contrast to QuerySet, is a nodeset selected in various ways on the network map.

327
Using NetMiner

The selected nodes itself (not the condition to select these nodes as was in the case of

QuerySet) with yellow color can be saved as Selection. The condition for selecting nodes on the

network map is not saved as Selection. You can refer to “Using Saved QuerySet and Selection

in Filter” for applying the saved Selection to each Module for running process.

Note that linkset cannot be saved as Selection. Only nodes on the network map can be saved as

Selection. You can show and hide linkset on network map using Weight Threshold Control item

in Select Conrol panel. But if you are to save the linkset on the network map, you must save the

condition for showing the links on the network map as QuerySet.

You can check the saved QuerySet and Selection in Current Workfile as follows.

 Saving Selected Nodes as a Selection

You can save selected nodes on the network map as a Selection. The saved Selection can be

used for running Process Module in Filter Control item and be used to select nodes on network

map by loading it. If a complex condition is required to select nodes on network map, the saved

Selection will be helpful to select nodes conveniently. This topic covers saving selected nodes

as Selection by example. Using Selection is covered in “Using Saved QuerySet and Selection in

Filter” and “Loading Selection to Current Network Map”.

328
 III.Task

For example, we will select nodes on the “Trust” network map with following condition in

Condition Control item in Select Control panel.

"Gender" == "male" and "Age" >= 30.0

Click Run Query button for selecting on the network map.

Now you can see the selected nodes and links on the network map with yellow color. Click

Save button in the Selection Control item, you can see a dialogue box for saving the selected

nodes. We have named this Selection as “Gender = male and age >= 30”.

You can see the following message box if you click OK button.

329
Using NetMiner

You can see the saved selection as a new item in Selections in Current Workfile as follows.

Now you can use the saved Selection in Filter for running Process Module and load the

selection to current network map.

 Saving Link Threshold as a QuerySet

You can show and hide the links on the network map using Link Threshold Control item. You

can save the condition for showing links on the network map as a QuerySet.

Step1: Show links the weight of which is over 5 using Link Threshold slide bar for the “Trust”
network map as follows.

330
 III.Task

Step 2: Click Save Threshold as QuerySet button for saving the condition for showing links on
the network map. You can see the following dialogue box for saving QuerySet. We named it as

“Trust: Weight >=4”.

Step 3: You can see the saved QuerySet in QuerySet in Current Workfile as follows.

331
Using NetMiner

 Using Saved QuerySet and Selection in Filter

The Saved QuerySet and Selection can be used in Process Modules. You can load the saved

QuerySet and Selection in Filter Control item in Process Control panel. We will use example

that was used in the previous topic “Saving Selected Nodes as a Selection” and “Saving Weight

Threshold as a QuerySet”

Step 1. Create a Degree Centrality Session by selecting main menu “Analyze >> Centraltiy >>
Degree”. Expand the Filter Control item. You can refer to the full description about Filter in

“Process Handling – Handling Filter”.

332
 III.Task

Step 2. You can launch a dialogue box for selecting QuerySet and Selection by clicking Load
button. You can see the saved QuerySets and Selections in the dialogue box. QuerySet and

Selection are classified in Type column. Session in Type column represents that the item is

applied to a Session.

Step 3. We will load a QuerySet “Trust:Weight >=4”. Select the item at the Name column and
click OK. You can see that the numbers attached to “Trust” data item changed. Now the

selected QuerySet is applied to the Filter.

Select “Trust” as the input data in Input Control item and click Run Process button. You can see

that the Degree Centrality analysis was performed only with the ‘filtered’ links. If you select

the [M] Spring tab, you can see the following disconnected network map due to the exclusion

of links.

333
Using NetMiner

Step 4. Now we are going to run Degree Centrality process again with another filter. We are
going to load a Selection to Filter. Click load button for launching the dialogue box for

selecting QuerySet and Selection and select “Gender = male and Age >=30”. You can see that

the numbers attached to Main Nodeset data item changed as follows.

Select “Trust” as the input data in Input Control item and click Run Process button. You can see

that the Degree Centrality analysis was performed only with the ‘filtered’ nodes. If you select

the [M] Spring tab, you can see the following network map composed of only the ‘filtered’

nodes.

334
 III.Task

 “Run Process” with Selected Nodes

If you are to run again the current Process with selected nodes, you need to save the selected

nodes as Selection and load it as Filter in Filter Control item in Process Control panel. And

then finally you need to click the Run Process button for running process. We have covered

these processes one by one in the previous topics. But you can run again the current Process

very easily using “Run Process with Selected Nodes” in mouse right click menu.

If you saved selected nodes as Selection, you can use that in another Process. But “Run Process

with Selected Nodes” can be used only in the current Process. It is very helpful to explore the

current network quickly using mouse.

If you select “Run Process with Selected Nodes” on the network map which is generated in the

‘Analyze’ menu, the default output is not the ‘Main Report’ but the network map on which you

selected “Run Process with Selected Nodes”.

Example 1. In Spring 2D Session created by selecting “Visualize >> Spring >> 2D”, select
“Trust” network and click Run Layout button. On the network map, select some nodes by

mouse dragging and select “Run Process with Selected Nodes” in the mouse right click menu

as follows.

335
Using NetMiner

You can quickly run again the current Spring 2D Session and draw a new network map with the

selected nodes as follows.

Example 2. Create Degree Centrality Session by selecting main menu “Analyze >>
Centrality >> Degree” and select 1-mode Network data “Trust”. Click Run Process button.

Select [M] Concentric, you can see a concentric map as follows.

336
 III.Task

Now select some nodes on the centric map and select “Run Process with Selected Nodes” in the

mouse right click menu. You can check that a new analysis results are generated with the

selected nodes as follows.

 Saving Selection as a New Workfile

You can create a new Workfile with the selected nodes on the network map.

337
Using NetMiner

Step 1: Select nodes on the network map using mouse as follows and click Save Selection as a
New Workfile button.

You can see a message box as follows.

You can see a new Workfile is expanded as follows. You can check that the number of Main

Nodeset of the expanded Workfile is changed.

 Loading Selection to Current Network Map

You can select nodes on the network map by loading the saved Selection. Click Load button in

the Selection Control item, you can see the dialogue box for selecting a Selection as follows.

338
 III.Task

Select a Selection and click OK button, and then you can see that some nodes on the network

map are selected as follows. You can also select nodes in the Current Workfile directly. Select

“Load” in the mouse right click menu at the selected Selection in Current Workfile. In this case

the network map must be activated on the Output panel.

 Styling Selected Nodes and Links

You can give various styles to the selected nodes and links on the network map. That will help

you to understand the structure of the network map intuitively. Full description about styling

nodes and links are covered in “Styling nodes and links”.

339
Using NetMiner

 Visualizing Neighbors of a Selected Node

You can visualize the neighbors of a selected node in a new sub window. The nodes on the new

sub window maintain the styles which were given on the original network map.

On “Trust” network map, select a node and select “Visualize Neighbor” in the mouse right click

menu as follows. You can see a new sub window showing the neighbors of the selected node.

The links between the alters (neighbor nodes) are shown. You can zoom in(out) using the

navigation bar at the right side.

 Checking Information About a Selected Node(Link)

You can check properties of a selected node or link.

• Properties of a Node

Select a node on the network map and select “Node Properties” in the mouse right click menu.

You can see an information message box as follows. Network Property represents the Label, In-

340
 III.Task

Degree, Out-Degree and Egonet Size of the selected node. Attribute Property represents the

values of the attribute item of the selected node.

• Properties of a Link

Select a link on the network map and select “Link Properties” in the mouse right click menu.

You can see an information message box as follows. You can see the weight and the attribute

value of the selected link.

In the above example screenshot, both the “Trust” and “Knowledge” network are selected as

input data. You can see that the selected link exists in “Trust” and “Knowledge” network at the

same time. You can see the weight and the attribute values of the selected link.

341
Using NetMiner

7. Inspecting Analysis Results in Network Map

 Overview

When network map(s) are generated as outputs in a Session with other analysis results as

Output Tables and Main Reports, you can inspect the analysis results on the network map using

Inspect Control panel. Inspect Control panel contains module-specific Control items to support

interactive and exploratory analysis. You can maximize the exploratory analysis efficiency

using Inspect Control panel with various select-and-run functions that we covered in the

previous sections.

The core factors of Inspect Control panel are Module- specific 1) Network Map Layout, 2)

Styles of nodes and links and 3) Control items.

Each Analyze Module was designed to generate most suitable network map for that Module.

And the styles of nodes and links on the network map are designed to represent the analysis

result intuitively when you inspect the analysis result in the network map using the Control

items which were also designed to fit for the analysis process using the analysis result.

We will try inspecting network map using some examples in the following topics.

 Examples

Analyze >> Connection >> Shortest Path

Step 1: Create Shortest Path Session by selecting main menu “Analyze >> Connection >>
Shortest Path” and select 1-mode Network data “Trust”. Click Run Process button, you can see

three outputs: [R] Main, [T] Distance Matrix and [M] Spring.

342
 III.Task

Step 2: Select [M] Spring in the Output panel and Select Inspect Control panel. You can see
two Control items: k-Shortest Path and k-Neighbor. Select CY163 as Source Node and CY013

as Target Node. You can search nodes easily if you fill the text box with numbers concerned. In

this example if you fill the text box with ‘63’, you can select CY163 node easily as follows.

Click Submit button, and then you will see the Shortest Path between two selected nodes on the

network map. You can also inspect the various paths between two nodes by selecting Distance.

In this example, there are three paths as you see.

Analyze >> Cohesion >> Component

Step 1: Create Component Session by selecting main menu “Analyze >> Cohesion >>
Component” and select 1-mode Network data “Friendship”. Select ‘2’ as the Minimum Size of

Component and “Weak Component” as the Component Type in Process panel and click Run

343
Using NetMiner

Process button. You can see three outputs: [R] Main, [T] Component Partition Vector, [M]

Clustered.

Step 2: Select [M] Clustered in the Output panel and select Inspect Control item. You can see a
Control item named Select Component. You can select a Component in the radio button list.

You can select whether or not to show the red dotted box that represents a component on the

network map by Show Border check box. The color of nodes in the selected Component

changes to red.

Step 3. Select a Component using mouse drag and select “Run Process with Selected Nodes” in
the mouse right click menu on the network map. You can extract the selected nodes and run the

Component Session again with the selected nodes as follows. We have selected only one

component, so all the component partition vector of the nodes are the same.

Step 4: Now select “Strong Component” as the Component Type in the Process Control panel.
And then click Run Process button. You can see that one Weak Component is separated into

many Strong Components. Select [M] Clustered, you can see the clustered network map as

344
 III.Task

follows. You can refer to the Module Reference for the details about Component Module.

Analyze >> Cohesion >> Community >> Betweenness

Step 1: Create Betweenness Community Session by selecting main menu “Analyze >>
Cohesion >> Community >> Betweenness” and select 1-mode Network data “knowledge”. If

you click Run Process button, five outputs are generated: [R] Main, [T] Community Cluster

Matrix, [T] Permutation Vector, [C] Dendrogram and [M] Clustered. Full descriptions are

covered in “Module Reference”. Select [M] Clustered in the Output panel and select Inspect

Control panel.

Step 2: You can see one Control item: “Community”. You can select the criteria for separating
the network map into several cohesive clusters based on the Fusion Level and Best Cut value.

Generally, the optimal separation criterion is Best Cut. So you can select the Fusion Level with

the highest Best Cut value. But you can inspect several cases to find best separation, which can

be easily done just by mouse clicking.

345
Using NetMiner

* Clustered – Eades layout is applied to the above screenshot.

Analyze >> Centrality >> Degree

Step 1: Create Degree Centrality Session by selecting main menu “Analyze >> Centrality >>
Degree”. Select 1-mode Network data “Trust” and click Run Process button. You can see four

outputs: [R] Main, [T] Degree Centrality Vector, [M] Spring and [M] Concentric.

Step 2: Select [M] Concentric in the Output panel and select Inspect Control panel. You can
see the Choose Direction Control item. You can select In-Degree or Out-Degree.

346
 III.Task

Step 3: Now we are going to select some nodes with specific condition and run process with
the selected nodes. We are going to select nodes who are male and over thirties. Move to Select

Control Panel and write the condition in the Selection Control item as follows:

“gender = male and Age>=30”

Click Run Query button, you can see the selected nodes on the network map. Now select “Run

Process with Selected Nodes” in the mouse right click menu on the network map.

347
Using NetMiner

Step4: You can run again the Degree Centrality Session again with the selected nodes. You can
see that the new concentric map is generated as follows.

348
 III.Task

8. Recording Visual Exploration

 Overview

NetMiner supports recording and editing the network map

exploration. You can record the network map exploration using

Video Control panel. So, this feature works only in sessions that

produce the map report.

If you finish the first recording process with clicking 'Record Stop' or the map already has a

video object, the 'time slide bar' is appeared on the map. ; You can play, stop, pause the video or

adjust segment ranges by using this slide bar.

349
Using NetMiner

1: play/pause button

2: stop & rewind button

3: start indicator of the selected range to play/edit

4: end indicator of the selected range to play/edit

5: navigation pointer

6: elapsed time

7: total running time

8: The ranges which have actual moving events are colored with light blue color.

 Recording a new video

Step 1: Create a visualize session and draw a network map you want to explore.
* In this example, we use the map created by Visualize>Sping3D.

Step 2: Click the Video control panel and start recording with clicking 'Record New'.

350
 III.Task

Step 3: Perform your exploration!

* Operations that can be recorded are changes of zoom, style, selection(highlighting), layout,

panning, flipping(2D), position of selected nodes(2D), and rotating(3D).

Step 4: Stop recording with clicking 'Record Stop'

 Recording a video to insert

Step 1: Move the navigation pointer on the time slide bar to insert.

Step 2: Start recording with clicking 'Record Insert'.

Step 3: Perform your exploration!

Step 4: Stop recording with clicking 'Record Stop'

 Selecting Range

All playing and editing operations are executed within the selected range.

You can specify a range of the video by adjusting start/end indicator located on the both sides

of time slide (yellow triangles).

351
Using NetMiner

Also, you can select a range according to segment units which are separated by color

(gray/light blue).

Select a range according to segment unit : Left-click with Shift key

Expand a range according to segment unit : Left-click with CTRL key

 Playing video

After finishing the first recording process with clicking 'Record Stop', you can see the 'time

slide bar' on the map.

Step 1: Adjust the range that you want to play.

Step 2: Set the playing start point with moving the navigation pointer

* If the navigation pointer is on the end of range, it is possible to play from the beginning of the

range by clicking the play button without need to rewind.

Step 3: Click the play button and watch video!

* During the video playback, if the case of the following events occur, the video

playing automatically pauses to prevent recording of undesired actions.

(1) The mouse pointer is out from the 'Output Panel' region.

(2) The click events except clicking the stop button occur.

352
 III.Task

 Editing video

You can edit the video after selecting range. Possible editing operations are as follows.

Change Speed : You can change play speed of the selected range by adjusting whole play time
value or ratio.

Dissolve : Connect recording actions more smoothly based on the beginning and end images in
the selected range.

Delete : Delete selected range of video

You can access the edit menus mentioned above by right-click on the time slide bar.

If the result of editing is not satisfied, you can undo/redo the previous editing step.

353
Using NetMiner

Appendix
A. Packages

B. Scalability and Performance

354
 Appendix

Appendix A. Packages

Basic
Advanced analysis
Exploration
Advanced visualization
Statistics
Options
Mining
Tools
Script
Semantic Network analysis

Updated full descriptions for each package are on the web site:

http://www.netminer.com >> Product >> Edition&Package.

355
Using NetMiner

Appendix B. Scalability and Performance

When we say NetMiner supports ‘Large Network Data’, what does it mean? We explore the

exact meaning of “NetMiner supports large network data analysis” through four questions and

answers below.

1. Do all licenses of NetMiner support ‘Large Network Data’?

No. There are four different levels of license (Tiny, Small, Medium, and Large)

with respect to the data size for NetMiner to handle. In order to handle ‘Large

Network Data,’ of maximum of 100,000 nodes, the appropriate level of NetMiner

license should be purchased.

2. If I purchased the appropriate level of NetMiner license in order to meet data

size of my dataset to be handled, would NetMiner work on the large-scale

dataset no matter what?

No. The level of license with restriction on data size just checks the maximum

nodes to be imported into NetMiner. However, the overall size of network data is

defined not only by the number of nodes but also by the number of links. In order
to set the upper limit of link size, NetMiner defines an operational definition of

‘Sparse Network’ as the network of ‘m ≤ 10 x n’, where ‘n’ is the number of nodes

and ‘m’ is the number of links. In other words, the number of links should be

smaller than ten times the node size and the maximum size of links is 1,000,000.

(NetMiner can run data with more than 1,000,000 links, but it is just not officially

guaranteed and

tested.)

3. If I purchased the appropriate level of NetMiner license in order to meet data

356
 Appendix

size of dataset to be handled and the dataset satisfies our operational large

sparse network definition, will all NetMiner algorithms operate efficiently

regardless of its time complexity?

No. NetMiner guarantees for the performance of algorithms with subquadratic

time complexity where running time f(n) is under o(n2). Test results of

subquadratic algorithms of NetMiner are listed below. Although over-quadratic

algorithms can run in NetMiner, it is not as fast and smooth processing of ‘Large

Network Data’ as compared to subquadratic algorithms. We will provide the

performance test results for over-quadratic algorithms later.

4. If I purchased the appropriate level of NetMiner license in order to meet data

size of dataset to be handled and the dataset satisfies our operational large

sparse network definition, and if I run subquadratic algorithms of NetMiner,

will it be guaranteed that those run with high performance?

No. Although your system satisfies the minimum requirements to install NetMiner,

the system performance such as RAM size and CPU speed inevitably affects the

NetMiner performance. Therefore, although NetMiner can run with lower-level

system specifications, it is recommended to have a large size of RAM and high

speed of CPU for better performance.

To sum it up, in order to get the benefit of NetMiner's handling large-scale network data, i)

users need to have appropriate level of license and, ii) the network data is sparse (m ≤ 10 x n)

and, iii) users are running subquadratic algorithms and, iv) NetMiner is installed in a

computer with a large size of RAM and high speed of CPU.

The contents below are the results of performance test using ‘Large but Sparse Network Data’

under a normal computing environment.

357
Using NetMiner

- The results of performance test of NetMiner

The contents of this document are regarding the performance of NetMiner under ‘Large

Network Data’ condition. In spite of the same hardware condition, different result can be

occurred.

1. Data for test

Single network data with properties below was used for test.

1. Directed network

2. Weighted network

3. Multiple links are allowed between two nodes

4. No Link Attributes nor Node Attributes were loaded unless analysis requires it(for

example, Blockmodeling or Node Extract)

# of Nodes 100,000

# of Links 1,012,705

Average Degree 10.127

Network Density 0.01%

2. Computing Environments

NetMiner can under following environments, where its 800MB memory was already

occupied with NetMiner and other applications.

CPU Intel○
R Core 2 6400, 2.13GHz

RAM 2GB

HDD 149GB, 7200rpm

358
 Appendix

OS Microsoft Windows XP○

R SP2

3. Selecting processing modules to test

All subquadratic algorithms in ‘Transform’ and ‘Analyze’ menu are tested.

4. Measure Elapsed time

Only time for loading and running algorithms(from starting command to generating report

pages) was measured, and time for generation of report pages was not included. ‘Elapsed Time’

is averaged over 5 runs.

5. Results

Menu Time Elapsed

Selection
nd rd
Top-Level 2 Level 3 Level Complexity Time

Neighbor Degree O(m) < 10 sec.

Weak O(m) < 10 sec.

Component
Strong O(m) < 10 sec.
Cohesion
Bi-
O(m) < 10 sec.
Component

Degree O(m) < 10 sec.

Cetrality
Analyze Coreness O(m) < 10 sec.

Number of
O(m) < 10 sec.
Links

Density O(m) < 10 sec.

Properties Network
Avg. Degree O(m) < 10 sec.

Inclusiveness O(m) < 10 sec.

Reciprocity O(m) < 10 sec.

359
Using NetMiner

Connectedness O(m) < 10 sec.

Efficiency O(m) < 10 sec.

Position Blockmodel O(m) < 30 sec.

Symmetrize O(m) < 10 sec.

Direction
Transpose O(m) < 10 sec.

Dichotomize O(m) < 10 sec.

Reverse O(m) < 10 sec.

Weight Normalize O(m) < 10 sec.

Recode O(m) < 10 sec.

Transform
Extract O(m) < 10 sec.

Ego
O(m) < 10 sec.
Node Set Networks

Reorder O(n log n) < 10 sec.

Extract O(n log n) < 10 sec.

Link Set
Incidence O(m) < 10 sec.

(n: the number of nodes, m: the number of links)

360
“Unleashing Hidden Power of Networks”

http://www.netminer.com

Tel: +82-31-739-8352

Fax: +82-31-739-8354

Email. netminer@cyram.com

361