Introduction To Data Mining

Data Mining
Session 1 – Main Theme

Introduction to Data Mining
Dr. Jean-Claude Franchitti
New York University

Computer Science Department
Courant Institute of Mathematical Sciences
Adapted from course textbook resources

Data Mining Concepts and Techniques (2nd Edition)
Jiawei Han and Micheline Kamber
Agenda
11 Instructor
Instructor and
and Course
Course Introduction
Introduction
22 Introduction
Introduction to
to Data
Data Mining
Mining
33 Summary
Summary and
and Conclusion
Conclusion
2
Who am I?
- Profile -
¾ 27 years of experience in the Information Technology Industry, including thirteen years of experience
working for leading IT consulting firms such as Computer Sciences Corporation
¾ PhD in Computer Science from University of Colorado at Boulder
¾ Past CEO and CTO
¾ Held senior management and technical leadership roles in many large IT Strategy and Modernization
projects for fortune 500 corporations in the insurance, banking, investment banking, pharmaceutical, retail,
and information management industries
¾ Contributed to several high-profile ARPA and NSF research projects
¾ Played an active role as a member of the OMG, ODMG, and X3H2 standards committees and as a
Professor of Computer Science at Columbia initially and New York University since 1997
¾ Proven record of delivering business solutions on time and on budget
¾ Original designer and developer of jcrew.com and the suite of products now known as IBM InfoSphere
DataStage
¾ Creator of the Enterprise Architecture Management Framework (EAMF) and main contributor to the creation
of various maturity assessment methodology
¾ Developed partnerships between several companies and New York University to incubate new
methodologies (e.g., EA maturity assessment methodology developed in Fall 2008), develop proof of
concept software, recruit skilled graduates, and increase the companies’ visibility
3
How to reach me?
Come on…what else

did you expect? Cell (212) 203-5004
Email jcf@cs.nyu.edu
AIM, Y! IM, ICQ jcf2_2003
MSN IM jcf2_2003@yahoo.com
LinkedIn http://www.linkedin.com/in/jcfranchitti
Woo hoo…find the word
of the day… Twitter http://twitter.com/jcfranchitti
Skype jcf2_2003@yahoo.com
4
What is the class about?
Course description and syllabus:

» http://www.nyu.edu/classes/jcf/g22.3033-002/
» http://www.cs.nyu.edu/courses/spring10/G22.3033-002/index.html
Textbooks:
» Data Mining: Concepts and Techniques (2nd Edition)
Jiawei Han, Micheline Kamber
Morgan Kaufmann
ISBN-10: 1-55860-901-6, ISBN-13: 978-1-55860-901-3, (2006)
» Microsoft SQL Server 2008 Analysis Services Step by Step
Scott Cameron
Microsoft Press
ISBN-10: 0-73562-620-0, ISBN-13: 978-0-73562-620-31 1st Edition (04/15/09)
Icons / Metaphors
Information
Common Realization
Knowledge/Competency Pattern
Governance
Alignment
Solution Approach
66
Agenda
11 Instructor
Instructor and
and Course
Course Introduction
Introduction
22 Introduction
Introduction to
to Data
Data Mining
Mining
33 Summary
Summary and
and Conclusion
Conclusion
Introduction to Data Mining - Sub-Topics

Why Data Mining?
» Data Mining: A Natural Evolution of Science and Technology
What Is Data Mining?
» Data Mining: Essential in a Knowledge Discovery Process
» Data Mining: A Confluence of Multiple Disciplines
A Multi-Dimensional View of Data Mining
» Knowledge to Be Mined
» Data to Be Mined
» Technology Utilized
» Applications Adapted
Data Mining Functionalities: What Kinds of Patterns Can Be Mined?
» Generalization
» Mining Frequent Patterns, Associations, and Correlations
» Classification
» Cluster Analysis
» Outlier Analysis
Data mining: On What Kinds of Data?
Time and Ordering: Sequential Pattern, Trend and Evolution Analysis
Structure and Network Analysis
Evaluation of knowledge
Applications of Data Mining
Major Challenges in Data Mining
A Brief History of Data Mining and Data Mining Society
8
Evolution of Sciences
Before 1600, empirical science

1600-1950s, theoretical science
» Each discipline has grown a theoretical component. Theoretical models often
motivate experiments and generalize our understanding.
1950s-1990s, computational science
» Over the last 50 years, most disciplines have grown a third, computational branch
(e.g. empirical, theoretical, and computational ecology, or physics, or linguistics.)
» Computational Science traditionally meant simulation. It grew out of our inability to
find closed-form solutions for complex mathematical models.
1990-now, data science
» The flood of data from new scientific instruments and simulations
» The ability to economically store and manage petabytes of data online
» The Internet and computing Grid that makes all these archives universally
accessible
» Scientific info. management, acquisition, organization, query, and visualization tasks
scale almost linearly with data volumes. Data mining is a major new challenge!
Jim Gray and Alex Szalay, The World Wide Telescope: An Archetype for Online Science,
Comm. ACM, 45(11): 50-54, Nov. 2002
Evolution of Database Technology (1/2)
1960s:
» Data collection, database creation, IMS and network DBMS
1970s:
» Relational data model, relational DBMS implementation
1980s:
» RDBMS, advanced data models (extended-relational, OO, deductive,
etc.)
» Application-oriented DBMS (spatial, scientific, engineering, etc.)
1990s:
» Data mining, data warehousing, multimedia databases, and Web
databases
2000s
» Stream data management and mining
» Data mining and its applications
» Web technology (XML, data integration) and global information systems
10
Evolution of Database Technology (2/2)
11
Why Data Mining? (1/2)
The Explosive Growth of Data: from terabytes to petabytes

» Data collection and data availability
• Automated data collection tools, database systems, Web,
computerized society
» Major sources of abundant data

• Business: Web, e-commerce, transactions, stocks, …
• Science: Remote sensing, bioinformatics, scientific simulation,
…
• Society and everyone: news, digital cameras, YouTube
We are drowning in data, but starving for knowledge!

“Necessity is the mother of invention”—Data mining—
Automated analysis of massive data sets
12
Why Data Mining? (2/2)
•Associations (e.g. linking purchase of pizza with

beer)
•Sequences (e.g. tying events together: marriage

and purchase of furniture)
•Classifications (e.g. recognizing patterns such as

the attributes of employees that are most likely to
quit)
•Forecasting (e.g. predicting buying habits of

customers based on past patterns) Expert
systems or small ML/statistical programs
13
What Can Data Mining Do?
Classification
» Classify credit applicants as low, medium, high risk
» Classify insurance claims as normal, suspicious
Estimation
» Estimate the probability of a direct mailing response
» Estimate the lifetime value of a customer
Prediction
» Predict which customers will leave within six months
» Predict the size of the balance that will be transferred by a
credit card prospect
Association
» Find out items customers are likely to buy together
» Find out what books to recommend to Amazon.com users
Clustering
» Difference from classification: classes are unknown!
14
Sample Data Mining Algorithms
Data Mining Algorithms
Online Analytical Discovery Driven Methods

Processing
Description Prediction
SQL Query Tools
Classification Regressions
Visualization
Decision Trees
Clustering
Association Neural Networks
Sequential Analysis
15
Why Data Mining?—Potential Applications
Data analysis and decision support

» Market analysis and management
• Target marketing, customer relationship management (CRM), market
basket analysis, cross selling, market segmentation
» Risk analysis and management
• Forecasting, customer retention, improved underwriting, quality
control, competitive analysis
» Fraud detection and detection of unusual patterns (outliers)
Other Applications
» Text mining (news group, email, documents) and Web mining
» Stream data mining
» Bioinformatics and bio-data analysis
16
Example 1: Market Analysis and Management
Where does the data come from?—Credit card transactions, loyalty cards, discount coupons,
customer complaint calls, plus (public) lifestyle studies
Target marketing
» Find clusters of “model” customers who share the same characteristics: interest, income level, spending habits,
etc.
» Determine customer purchasing patterns over time
Direct Marketing
» Identify which prospects should be included in a mailing list
Market segmentation
» identify common characteristics of customers who buy same products
Market Basket Analysis
» Identify what products are likely to be bought together
Cross-market analysis—Find associations/co-relations between product sales, & predict based on
such association
Customer profiling—What types of customers buy what products (clustering or classification)
Customer requirement analysis
» Identify the best products for different groups of customers
» Predict what factors will attract new customers
Provision of summary information
» Multidimensional summary reports
» Statistical summary information (data central tendency and variation)
17
Sample Market Basket Analysis
Association and sequence discovery

Principal concepts
Support or Prevalence: frequency that a particular
association appears in the database
Confidence: conditional predictability of B, given A
Example:
Total daily transactions: 1,000
Number which include “soda”: 500
Number which include “orange juice”: 800
Number which include “soda” and “orange juice”: 450
SUPPORT for “soda and orange juice” = 45% (450/1,000)
CONFIDENCE of “soda à orange juice” = 90% (450/500)
CONFIDENCE of “orange juice à soda” = 56% (450/800)
18
Example 2: Corporate Analysis & Risk Management
Finance planning and asset evaluation

» cash flow analysis and prediction
» contingent claim analysis to evaluate assets
» cross-sectional and time series analysis (financial-ratio, trend
analysis, etc.)
Resource planning
» summarize and compare the resources and spending
Competition
» monitor competitors and market directions
» group customers into classes and a class-based pricing
procedure
» set pricing strategy in a highly competitive market
19
Example 3: Fraud Detection & Mining Unusual Patterns
Approaches: Clustering & model construction for frauds, outlier analysis

Applications: Health care, retail, credit card service, telecomm.
» Auto insurance: ring of collisions, insurance Claims Analysis
• Discover patterns of fraudulent transactions
• Compare current transactions against those patterns
» Money laundering: suspicious monetary transactions
» Medical insurance
• Professional patients, ring of doctors, and ring of references
• Unnecessary or correlated screening tests
» Telecommunications: phone-call fraud
• Phone call model: destination of the call, duration, time of day or week.
Analyze patterns that deviate from an expected norm
» Retail industry
• Analysts estimate that 38% of retail shrink is due to dishonest employees
» Anti-terrorism
20
Other Applications (1/2)
Sports
» IBM Advanced Scout analyzed NBA game statistics (shots
blocked, assists, and fouls) to gain competitive advantage for
New York Knicks and Miami Heat
Astronomy
» JPL and the Palomar Observatory discovered 22 quasars with
the help of data mining
Internet Web Surf-Aid
» IBM Surf-Aid applies data mining algorithms to Web access
logs for market-related pages to discover customer preference
and behavior pages, analyzing effectiveness of Web marketing,
improving Web site organization, etc.
21
Example: Amazon.com book recommendations
Example: Identify books to recommend to customers
Company keeps log of past customer purchases

Represent each customer as a vector whose
components are the past purchases
Define a “distance” function for comparing customers
Based on this distance function, identify the
customer’s nearest neighbor set (NNS)
Identify books that have been purchased by a large
percentage of the nearest neighbor set but not by the
customer
Recommend these books to the customer as possible
next purchases
22
Introduction to Data Mining – Sub-Topics
Why Data Mining?
» Data to Be Mined
» Generalization
» Classification
» Cluster Analysis
» Outlier Analysis
23
Data mining (knowledge discovery from data)

» Extraction of interesting (non-trivial, implicit, previously unknown
and potentially useful) patterns or knowledge from huge amount
of data
» Data mining: a misnomer?
Alternative names
» Knowledge discovery (mining) in databases (KDD), knowledge
extraction, data/pattern analysis, data archeology, data
dredging, information harvesting, business intelligence, etc.
Watch out: Is everything “data mining”?
» Simple search and query processing
» (Deductive) expert systems
24
Knowledge Discovery (KDD) Process
» Data mining—core of
knowledge discovery Pattern Evaluation
process and Presentation
Data Mining
Task-relevant Data
Selection
Data Warehouse
Data Cleansing
Data Integration
Databases
25
Example: A Web Mining Framework (1/2)
Web mining usually involves

» Data cleaning
» Data integration from multiple sources
» Warehousing the data
» Data cube construction
» Data selection for data mining
» Data mining
» Presentation of the mining results
» Patterns and knowledge to be used or stored into knowledge-
base
26
Example: A Web Mining Framework (2/2)
27
KDD Process: Several Key Steps
Learning the application domain

» relevant prior knowledge and goals of application
Creating a target data set: data selection
Data cleaning and preprocessing: (may take 60% of effort!)
Data reduction and transformation
» Find useful features, dimensionality/variable reduction, invariant
representation
Choosing functions of data mining
» summarization, classification, regression, association, clustering
Choosing the mining algorithm(s)
Data mining: search for patterns of interest
Pattern evaluation and knowledge presentation
» visualization, transformation, removing redundant patterns, etc.
Use of discovered knowledge
28
Are All the “Discovered” Patterns Interesting?
Data mining may generate thousands of patterns: Not all of them are
interesting
» Suggested approach: Human-centered, query-based, focused mining
Interestingness measures
» A pattern is interesting if it is easily understood by humans, valid on new
or test data with some degree of certainty, potentially useful, novel, or
validates some hypothesis that a user seeks to confirm
Objective vs. subjective interestingness measures
» Objective: based on statistics and structures of patterns, e.g., support,
confidence, etc.
» Subjective: based on user’s belief in the data, e.g., unexpectedness,
novelty, actionability, etc.
29
Find All and Only Interesting Patterns?
Find all the interesting patterns: Completeness

» Can a data mining system find all the interesting patterns? Do we
need to find all of the interesting patterns?
» Heuristic vs. exhaustive search
» Association vs. classification vs. clustering
Search for only interesting patterns: An optimization problem
» Can a data mining system find only the interesting patterns?
» Approaches
• First general all the patterns and then filter out the uninteresting ones
• Generate only the interesting patterns—mining query optimization
30
Other Pattern Mining Issues
Precise patterns vs. approximate patterns

» Association and correlation mining: possible find sets of precise
patterns
• But approximate patterns can be more compact and sufficient
• How to find high quality approximate patterns??
» Gene sequence mining: approximate patterns are inherent
• How to derive efficient approximate pattern mining algorithms??
Constrained vs. non-constrained patterns
» Why constraint-based mining?
» What are the possible kinds of constraints? How to push
constraints into the mining process?
31
Data Mining and Business Intelligence
Increasing potential
to support
business decisions End User
Decision
Making
Data Presentation Business

Analyst
Visualization Techniques
Data Mining Data
Information Discovery Analyst
Data Exploration
Statistical Summary, Querying, and Reporting
Data Preprocessing/Integration, Data Warehouses

DBA
Data Sources
Paper, Files, Web documents, Scientific experiments, Database Systems
32
Example: Mining vs. Data Exploration
Business intelligence view

» Warehouse, data cube, reporting but not much mining
Business objects vs. data mining tools
Supply chain example: tools
Data presentation
Exploration
33
KDD Process: A Typical View from ML and Statistics
Input Data Data Pre- Data Post-

Processing Mining Processing
Data integration Pattern discovery Pattern evaluation

Normalization Association & correlation Pattern selection
Feature selection Classification Pattern interpretation
Clustering
Dimension reduction Pattern visualization
Outlier analysis
…………
This is a view from typical machine learning and statistics
communities
34
Example: Medical Data Mining
Health care & medical data mining – often

adopted such a view in statistics and machine
learning
Preprocessing of the data (including feature
extraction and dimension reduction)
Classification or/and clustering processes
Post-processing for presentation
35
Data Mining: Confluence of Multiple Disciplines
Database
Technology Statistics
Machine Visualization
Learning Data Mining
Pattern
Recognition Other
Algorithm Disciplines
36
Why Data Mining?
» Data to Be Mined
» Generalization
» Classification
» Cluster Analysis
» Outlier Analysis
37
Multi-Dimensional View of Data Mining
Data to be mined
» Relational, data warehouse, transactional, stream, object-
oriented/relational, active, spatial, time-series, text, multi-media,
heterogeneous, legacy, WWW
Knowledge to be mined
» Characterization, discrimination, association, classification,
clustering, trend/deviation, outlier analysis, etc.
» Multiple/integrated functions and mining at multiple levels
Techniques utilized
» Database-oriented, data warehouse (OLAP), machine learning,
statistics, visualization, etc.
Applications adapted
» Retail, telecommunication, banking, fraud analysis, bio-data
mining, stock market analysis, text mining, Web mining, etc.
38
Why Confluence of Multiple Disciplines?
Tremendous amount of data

» Algorithms must be highly scalable to handle such as tera-bytes of
data
High-dimensionality of data
» Micro-array may have tens of thousands of dimensions
High complexity of data
» Data streams and sensor data
» Time-series data, temporal data, sequence data
» Structure data, graphs, social networks and multi-linked data
» Heterogeneous databases and legacy databases
» Spatial, spatiotemporal, multimedia, text and Web data
» Software programs, scientific simulations
New and sophisticated applications
39

Why Data Mining?
» Data to Be Mined
» Generalization
» Classification
» Cluster Analysis
» Outlier Analysis
40
Data Mining Function: (1) Generalization (1/4)
Information integration and data warehouse construction

» Data cleaning, transformation, integration, and
multidimensional data model
Data cube technology
» Scalable methods for computing (i.e., materializing)
multidimensional aggregates
» OLAP (online analytical processing)
Multidimensional concept description: Characterization
and discrimination
» Generalize, summarize, and contrast data
characteristics, e.g., dry vs. wet region
41
42
43
44
Data Mining Function: (2) Association and Correlation Analysis (1/2)
Frequent patterns (or frequent itemsets)

» What items are frequently purchased together in your
Walmart?
Association, correlation vs. causality
» A typical association rule
• Diaper Æ Beer [0.5%, 75%] (support, confidence)
» Are strongly associated items also strongly
correlated?
How to mine such patterns and rules efficiently in large
datasets?
How to use such patterns for classification, clustering,
and other applications?
45
Data Mining Function: (2) Association and Correlation Analysis (2/2)
46
Data Mining Function: (3) Classification (1/2)
Classification and label prediction

» Construct models (functions) based on some training examples
» Describe and distinguish classes or concepts for future prediction
• E.g., classify countries based on (climate), or classify cars based on
(gas mileage)
» Predict some unknown class labels
Typical methods
» Decision trees, naïve Bayesian classification, support vector
machines, neural networks, rule-based classification, pattern-
based classification, logistic regression, …
Typical applications:
» Credit card fraud detection, direct marketing, classifying stars,
diseases, web-pages, …
47
Data Mining Function: (3) Classification (2/2)
48
Data Mining Function: (4) Cluster Analysis
Unsupervised learning (i.e., Class label is unknown)

Group data to form new categories (i.e., clusters), e.g.,
cluster houses to find distribution patterns
Principle: Maximizing intra-class similarity & minimizing
interclass similarity
Many methods and applications
49
Data Mining Function: (5) Outlier Analysis
Outlier analysis
» Outlier: A data object that
does not comply with the
general behavior of the
data
» Noise or exception? ― One
person’s garbage could be
another person’s treasure
» Methods: by product of
clustering or regression
analysis, …
» Useful in fraud detection,
rare events analysis
50
Top-10 Most Popular DM Algorithms: 18 Identified Candidates (1/3)
Classification
» #1. C4.5: Quinlan, J. R. C4.5: Programs for Machine Learning.
Morgan Kaufmann., 1993.
» #2. CART: L. Breiman, J. Friedman, R. Olshen, and C. Stone.
Classification and Regression Trees. Wadsworth, 1984.
» #3. K Nearest Neighbours (kNN): Hastie, T. and Tibshirani, R.
1996. Discriminant Adaptive Nearest Neighbor Classification.
TPAMI. 18(6)
» #4. Naive Bayes Hand, D.J., Yu, K., 2001. Idiot's Bayes: Not So
Stupid After All? Internat. Statist. Rev. 69, 385-398.
Statistical Learning
» #5. SVM: Vapnik, V. N. 1995. The Nature of Statistical Learning
Theory. Springer-Verlag.
» #6. EM: McLachlan, G. and Peel, D. (2000). Finite Mixture
Models. J. Wiley, New York. Association Analysis
» #7. Apriori: Rakesh Agrawal and Ramakrishnan Srikant. Fast
Algorithms for Mining Association Rules. In VLDB '94.
» #8. FP-Tree: Han, J., Pei, J., and Yin, Y. 2000. Mining frequent
patterns without candidate generation. In SIGMOD '00.
51
The 18 Identified Candidates (2/3)
Link Mining
» #9. PageRank: Brin, S. and Page, L. 1998. The anatomy of a
large-scale hypertextual Web search engine. In WWW-7, 1998.
» #10. HITS: Kleinberg, J. M. 1998. Authoritative sources in a
hyperlinked environment. SODA, 1998.
Clustering
» #11. K-Means: MacQueen, J. B., Some methods for
classification and analysis of multivariate observations, in Proc.
5th Berkeley Symp. Mathematical Statistics and Probability,
1967.
» #12. BIRCH: Zhang, T., Ramakrishnan, R., and Livny, M. 1996.
BIRCH: an efficient data clustering method for very large
databases. In SIGMOD '96.
Bagging and Boosting
» #13. AdaBoost: Freund, Y. and Schapire, R. E. 1997. A decision-
theoretic generalization of on-line learning and an application to
boosting. J. Comput. Syst. Sci. 55, 1 (Aug. 1997), 119-139.
52
The 18 Identified Candidates (3/3)
Sequential Patterns
» #14. GSP: Srikant, R. and Agrawal, R. 1996. Mining Sequential
Patterns: Generalizations and Performance Improvements. In
Proceedings of the 5th International Conference on Extending
Database Technology, 1996.
» #15. PrefixSpan: J. Pei, J. Han, B. Mortazavi-Asl, H. Pinto, Q. Chen,
U. Dayal and M-C. Hsu. PrefixSpan: Mining Sequential Patterns
Efficiently by Prefix-Projected Pattern Growth. In ICDE '01.
Integrated Mining
» #16. CBA: Liu, B., Hsu, W. and Ma, Y. M. Integrating classification
and association rule mining. KDD-98.
Rough Sets
» #17. Finding reduct: Zdzislaw Pawlak, Rough Sets: Theoretical
Aspects of Reasoning about Data, Kluwer Academic Publishers,
Norwell, MA, 1992
Graph Mining
» #18. gSpan: Yan, X. and Han, J. 2002. gSpan: Graph-Based
Substructure Pattern Mining. In ICDM '02.
53
Top-10 Algorithm Finally Selected at ICDM’06
#1: C4.5 (61 votes)

#2: K-Means (60 votes)
#3: SVM (58 votes)
#4: Apriori (52 votes)
#5: EM (48 votes)
#6: PageRank (46 votes)
#7: AdaBoost (45 votes)
#7: kNN (45 votes)
#7: Naive Bayes (45 votes)
#10: CART (34 votes)
54
Why Data Mining?
» Data to Be Mined
» Generalization
» Classification
» Cluster Analysis
» Outlier Analysis
55
Data Mining: On What Kinds of Data?
Database-oriented data sets and applications

» Relational database, data warehouse, transactional database
Advanced data sets and advanced applications
» Data streams and sensor data
» Time-series data, temporal data, sequence data (incl. bio-
sequences)
» Structure data, graphs, social networks and multi-linked data
» Object-relational databases
» Heterogeneous databases and legacy databases
» Spatial data and spatiotemporal data
» Multimedia database
» Text databases
» The World-Wide Web
56
Why Data Mining?
» Data to Be Mined
» Generalization
» Classification
» Cluster Analysis
» Outlier Analysis
57
Sequence, trend and evolution analysis

» Trend, time-series, and deviation analysis: e.g.,
regression and value prediction
» Sequential pattern mining
• e.g., first buy digital camera, then buy large SD memory cards
» Periodicity analysis
» Motifs and biological sequence analysis
• Approximate and consecutive motifs
» Similarity-based analysis
Mining data streams
» Ordered, time-varying, potentially infinite, data streams
58
Why Data Mining?
» Data to Be Mined
» Generalization
» Classification
» Cluster Analysis
» Outlier Analysis
59
Graph mining
» Finding frequent subgraphs (e.g., chemical compounds), trees
(XML), substructures (web fragments)
Information network analysis
» Social networks: actors (objects, nodes) and relationships (edges)
• e.g., author networks in CS, terrorist networks
» Multiple heterogeneous networks
• A person could be multiple information networks: friends, family,
classmates, …
» Links carry a lot of semantic information: Link mining
Web mining
» Web is a big information network: from PageRank to Google
» Analysis of Web information networks
• Web community discovery, opinion mining, usage mining, …
60
Why Data Mining?
» Data to Be Mined
» Generalization
» Classification
» Cluster Analysis
» Outlier Analysis
61
Evaluation of Knowledge
Are all mined knowledge interesting?

» One can mine tremendous amount of “patterns” and knowledge
» Some may fit only certain dimension space (time, location, …)
» Some may not be representative, may be transient, …
Evaluation of mined knowledge → directly mine only
interesting knowledge?
» Descriptive vs. predictive
» Coverage
» Typicality vs. novelty
» Accuracy
» Timeliness
» …
62
Why Data Mining?
» Data to Be Mined
» Generalization
» Classification
» Cluster Analysis
» Outlier Analysis
63
Web page analysis: from web page classification, clustering to

PageRank & HITS algorithms
Collaborative analysis & recommender systems
Basket data analysis to targeted marketing
Biological and medical data analysis: classification, cluster analysis
(microarray data analysis), biological sequence analysis, biological
network analysis
Data mining and software engineering (e.g., IEEE Computer, Aug.
2009 issue)
From major dedicated data mining systems/tools (e.g., SAS, MS SQL-
Server Analysis Manager, Oracle Data Mining Tools) to invisible data
mining
64
Why Data Mining?
» Data to Be Mined
» Generalization
» Classification
» Cluster Analysis
» Outlier Analysis
Major Challenges in Data Mining – Additional Topics
65
Efficiency and scalability of data mining algorithms

Parallel, distributed, stream, and incremental mining methods
Handling high-dimensionality
Handling noise, uncertainty, and incompleteness of data
Incorporation of constraints, expert knowledge, and background
knowledge in data mining
Pattern evaluation and knowledge integration
Mining diverse and heterogeneous kinds of data: e.g., bioinformatics,
Web, software/system engineering, information networks
Application-oriented and domain-specific data mining
Invisible data mining (embedded in other functional modules)
Protection of security, integrity, and privacy in data mining
66
Focus Areas in Data Mining
Mining methodology
» Mining different kinds of knowledge from diverse data types, e.g., bio,
stream, Web
» Performance: efficiency, effectiveness, and scalability
» Pattern evaluation: the interestingness problem
» Incorporation of background knowledge
» Handling noise and incomplete data
» Parallel, distributed and incremental mining methods
» Integration of the discovered knowledge with existing one: knowledge
fusion
User interaction
» Data mining query languages and ad-hoc mining
» Expression and visualization of data mining results
» Interactive mining of knowledge at multiple levels of abstraction
Applications and social impacts
» Domain-specific data mining & invisible data mining
» Protection of data security, integrity, and privacy
67
Why Data Mining Query Language?
Automated vs. query-driven?

» Finding all the patterns autonomously in a database?—unrealistic
because the patterns could be too many but uninteresting
Data mining should be an interactive process
» User directs what to be mined
Users must be provided with a set of primitives to be used to
communicate with the data mining system
Incorporating these primitives in a data mining query language
» More flexible user interaction
» Foundation for design of graphical user interface
» Standardization of data mining industry and practice
68
Primitives that Define a Data Mining Task (1/2)
Task-relevant data
» Database or data warehouse name
» Database tables or data warehouse cubes
» Condition for data selection
» Relevant attributes or dimensions
» Data grouping criteria
Type of knowledge to be mined
» Characterization, discrimination, association, classification,
prediction, clustering, outlier analysis, other data mining tasks
Background knowledge
Pattern interestingness measurements
Visualization/presentation of discovered patterns
69
Primitives that Define a Data Mining Task (2/2)
70
Primitive 3: Background Knowledge (1/2)
A typical kind of background knowledge: Concept

hierarchies
Schema hierarchy
» E.g., street < city < province_or_state < country
Set-grouping hierarchy
» E.g., {20-39} = young, {40-59} = middle_aged
Operation-derived hierarchy
» email address: xyz@cs.nyu.edu
login-name < department < university < country
Rule-based hierarchy
» low_profit_margin (X) <= price(X, P1) and cost (X, P2)
and (P1 - P2) < $50
71
Primitive 3: Background Knowledge (2/2)
72
Primitive 4: Pattern Interestingness Measure
Simplicity
e.g., (association) rule length, (decision) tree size
Certainty
e.g., confidence, P(A|B) = #(A and B)/ #(B), classification
reliability or accuracy, certainty factor, rule strength, rule quality,
discriminating weight, etc.
Utility
potential usefulness, e.g., support (association), noise threshold
(description)
Novelty
not previously known, surprising (used to remove redundant
rules, e.g., Illinois vs. Champaign rule implication support ratio)
73
Primitive 5: Presentation of Discovered Patterns
Different backgrounds/usages may require different forms

of representation
» E.g., rules, tables, crosstabs, pie/bar chart, etc.
Concept hierarchy is also important
» Discovered knowledge might be more understandable
when represented at high level of abstraction
» Interactive drill up/down, pivoting, slicing and dicing
provide different perspectives to data
Different kinds of knowledge require different
representation: association, classification, clustering, etc.
74
DMQL—A Data Mining Query Language
Motivation
» A DMQL can provide the ability to support ad-hoc
and interactive data mining
» By providing a standardized language like SQL
• Hope to achieve a similar effect like that SQL has on
relational database
• Foundation for system development and evolution
• Facilitate information exchange, technology transfer,
commercialization and wide acceptance
Design
» DMQL is designed with the primitives described
earlier
75
An Example Query in DMQL
76
Other Data Mining Languages & Standardization Efforts
Association rule language specifications

» MSQL (Imielinski & Virmani’99)
» MineRule (Meo Psaila and Ceri’96)
» Query flocks based on Datalog syntax (Tsur et al’98)
OLEDB for DM (Microsoft’2000) and recently DMX (Microsoft
SQLServer 2005)
» Based on OLE, OLE DB, OLE DB for OLAP, C#
» Integrating DBMS, data warehouse and data mining
DMML (Data Mining Mark-up Language) by DMG (www.dmg.org)
» Providing a platform and process structure for effective data mining
» Emphasizing on deploying data mining technology to solve business
problems
77
Integration of Data Mining and Data Warehousing
Data mining systems, DBMS, Data warehouse systems coupling
» No coupling, loose-coupling, semi-tight-coupling, tight-coupling
On-line analytical mining data
» integration of mining and OLAP technologies
Interactive mining multi-level knowledge
» Necessity of mining knowledge and patterns at different levels of

abstraction by drilling/rolling, pivoting, slicing/dicing, etc.
Integration of multiple mining functions
» Characterized classification, first clustering and then association
78
Coupling Data Mining with DB/DW Systems
No coupling—flat file processing, not recommended

Loose coupling
» Fetching data from DB/DW
Semi-tight coupling—enhanced DM performance
» Provide efficient implement a few data mining primitives in a
DB/DW system, e.g., sorting, indexing, aggregation, histogram
analysis, multiway join, precomputation of some stat functions
Tight coupling—A uniform information processing
environment
» DM is smoothly integrated into a DB/DW system, mining query
is optimized based on mining query, indexing, query processing
methods, etc.
79
Architecture: Typical Data Mining System
Graphical User Interface
Pattern Evaluation
Data Mining Engine
Database or Data Knowledge-Base

Warehouse Server
data cleaning, integration, and selection
Data World-Wide Other Info

Database Warehouse Web Repositories
80
An OLAM Architecture
Mining Mining Layer4

query result User Interface
User GUI API
Layer3
OLAM OLAP
Engine Engine OLAP/OLA
M
Data Cube API
Layer2
MDD
MDDB
B Meta
Data
Filtering&Integratio Database API Filterin
n g Layer1
Data cleaning Data
Database Data
Data integration Warehous
s Repository
e
81

Why Data Mining?
» Data to Be Mined
» Generalization
» Classification
» Cluster Analysis
» Outlier Analysis
Major Challenges in Data Mining – Additional Topics
82
A Brief History of Data Mining Society
1989 IJCAI Workshop on Knowledge Discovery in Databases

» Knowledge Discovery in Databases (G. Piatetsky-Shapiro and W.
Frawley, 1991)
1991-1994 Workshops on Knowledge Discovery in Databases
» Advances in Knowledge Discovery and Data Mining (U. Fayyad,
G. Piatetsky-Shapiro, P. Smyth, and R. Uthurusamy, 1996)
1995-1998 International Conferences on Knowledge Discovery in
Databases and Data Mining (KDD’95-98)
» Journal of Data Mining and Knowledge Discovery (1997)
ACM SIGKDD conferences since 1998 and SIGKDD Explorations
More conferences on data mining
» PAKDD (1997), PKDD (1997), SIAM-Data Mining (2001), (IEEE)
ICDM (2001), etc.
ACM Transactions on KDD starting in 2007
83
Conferences and Journals on Data Mining
KDD Conferences Other related conferences

» ACM SIGKDD Int. Conf. on » ACM SIGMOD
Knowledge Discovery in
» VLDB
Databases and Data Mining
(KDD) » (IEEE) ICDE
» SIAM Data Mining Conf. » WWW, SIGIR
(SDM) » ICML, CVPR, NIPS
» (IEEE) Int. Conf. on Data Journals
Mining (ICDM)
» Data Mining and Knowledge
» Conf. on Principles and
Discovery (DAMI or DMKD)
practices of Knowledge
Discovery and Data Mining » IEEE Trans. On Knowledge
(PKDD) and Data Eng. (TKDE)
» Pacific-Asia Conf. on » KDD Explorations
Knowledge Discovery and » ACM Trans. on KDD
Data Mining (PAKDD)
84
Where to Find References? DBLP, CiteSeer, Google
Data mining and KDD (SIGKDD: CDROM)

» Conferences: ACM-SIGKDD, IEEE-ICDM, SIAM-DM, PKDD, PAKDD, etc.
» Journal: Data Mining and Knowledge Discovery, KDD Explorations, ACM TKDD
Database systems (SIGMOD: ACM SIGMOD Anthology—CD ROM)
» Conferences: ACM-SIGMOD, ACM-PODS, VLDB, IEEE-ICDE, EDBT, ICDT, DASFAA
» Journals: IEEE-TKDE, ACM-TODS/TOIS, JIIS, J. ACM, VLDB J., Info. Sys., etc.
AI & Machine Learning
» Conferences: Machine learning (ML), AAAI, IJCAI, COLT (Learning Theory), CVPR, NIPS,
etc.
» Journals: Machine Learning, Artificial Intelligence, Knowledge and Information Systems,
IEEE-PAMI, etc.
Web and IR
» Conferences: SIGIR, WWW, CIKM, etc.
» Journals: WWW: Internet and Web Information Systems,
Statistics
» Conferences: Joint Stat. Meeting, etc.
» Journals: Annals of statistics, etc.
Visualization
» Conference proceedings: CHI, ACM-SIGGraph, etc.
» Journals: IEEE Trans. visualization and computer graphics, etc.
85
Recommended Reference Books
S. Chakrabarti. Mining the Web: Statistical Analysis of Hypertex and Semi-Structured Data. Morgan
Kaufmann, 2002
R. O. Duda, P. E. Hart, and D. G. Stork, Pattern Classification, 2ed., Wiley-Interscience, 2000
T. Dasu and T. Johnson. Exploratory Data Mining and Data Cleaning. John Wiley & Sons, 2003
U. M. Fayyad, G. Piatetsky-Shapiro, P. Smyth, and R. Uthurusamy. Advances in Knowledge Discovery
and Data Mining. AAAI/MIT Press, 1996
U. Fayyad, G. Grinstein, and A. Wierse, Information Visualization in Data Mining and Knowledge
Discovery, Morgan Kaufmann, 2001
J. Han and M. Kamber. Data Mining: Concepts and Techniques. Morgan Kaufmann, 2nd ed., 2006
D. J. Hand, H. Mannila, and P. Smyth, Principles of Data Mining, MIT Press, 2001
T. Hastie, R. Tibshirani, and J. Friedman, The Elements of Statistical Learning: Data Mining,
Inference, and Prediction, Springer-Verlag, 2001
B. Liu, Web Data Mining, Springer 2006.
T. M. Mitchell, Machine Learning, McGraw Hill, 1997
G. Piatetsky-Shapiro and W. J. Frawley. Knowledge Discovery in Databases. AAAI/MIT Press, 1991
P.-N. Tan, M. Steinbach and V. Kumar, Introduction to Data Mining, Wiley, 2005
S. M. Weiss and N. Indurkhya, Predictive Data Mining, Morgan Kaufmann, 1998
I. H. Witten and E. Frank, Data Mining: Practical Machine Learning Tools and Techniques with Java
Implementations, Morgan Kaufmann, 2nd ed. 2005
86
Agenda
11 Instructor
Instructor and
and Course
Course Introduction
Introduction
22 Introduction
Introduction to
to Data
Data Mining
Mining
33 Summary
Summary and
and Conclusion
Conclusion
87
Summary
Data mining: Discovering interesting patterns from large amounts of

data
A natural evolution of database technology, in great demand, with
wide applications
A KDD process includes data cleaning, data integration, data
selection, transformation, data mining, pattern evaluation, and
knowledge presentation
Mining can be performed in a variety of information repositories
Data mining functionalities: characterization, discrimination,
association, classification, clustering, outlier and trend analysis, etc.
Data mining systems and architectures
Focus areas in data mining
88
Assignments & Readings
Readings
» Foreword/Preface and Chapter 1
Assignment #1
» Textbook Exercises 1.5, 1.7, 1.10, 1.11, 1.12, 1.15
89
Next Session: Data Preprocessing
90

Introduction To Data Mining

Uploaded by

Copyright:

Available Formats

Introduction To Data Mining

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Introduction To Data Mining

Uploaded by

Copyright:

Available Formats

What is the class about?

What is the class about?

What are some of the major topics that will be covered in the class?

What are some of the major topics that will be covered in the class?

Data Mining

Session 1 – Main Theme

Dr. Jean-Claude Franchitti

New York University

Adapted from course textbook resources

How to reach me?

Come on…what else

AIM, Y! IM, ICQ jcf2_2003

 Course description and syllabus:

Introduction to Data Mining - Sub-Topics

 Before 1600, empirical science

Evolution of Database Technology (1/2)

Why Data Mining? (1/2)

 The Explosive Growth of Data: from terabytes to petabytes

» Major sources of abundant data

 We are drowning in data, but starving for knowledge!

•Associations (e.g. linking purchase of pizza with

•Sequences (e.g. tying events together: marriage

•Classifications (e.g. recognizing patterns such as

•Forecasting (e.g. predicting buying habits of

What Can Data Mining Do?

Data Mining Algorithms

Online Analytical Discovery Driven Methods

Why Data Mining?—Potential Applications

 Data analysis and decision support

Sample Market Basket Analysis

 Association and sequence discovery

 Finance planning and asset evaluation

Example 3: Fraud Detection & Mining Unusual Patterns

 Approaches: Clustering & model construction for frauds, outlier analysis

Example: Amazon.com book recommendations

 Example: Identify books to recommend to customers

 Company keeps log of past customer purchases

What Is Data Mining?

 Data mining (knowledge discovery from data)

Example: A Web Mining Framework (1/2)

 Web mining usually involves

KDD Process: Several Key Steps

 Learning the application domain

Find All and Only Interesting Patterns?

 Find all the interesting patterns: Completeness

 Precise patterns vs. approximate patterns

Data Mining and Business Intelligence

Data Presentation Business

Data Preprocessing/Integration, Data Warehouses

 Business intelligence view

KDD Process: A Typical View from ML and Statistics

Input Data Data Pre- Data Post-

Data integration Pattern discovery Pattern evaluation

 Health care & medical data mining – often

Data Mining: Confluence of Multiple Disciplines

Multi-Dimensional View of Data Mining

 Tremendous amount of data

Introduction to Data Mining - Sub-Topics

 Information integration and data warehouse construction

Course description and syllabus:

Before 1600, empirical science

The Explosive Growth of Data: from terabytes to petabytes

We are drowning in data, but starving for knowledge!

Data analysis and decision support

Association and sequence discovery

Finance planning and asset evaluation

Approaches: Clustering & model construction for frauds, outlier analysis

Example: Identify books to recommend to customers

Company keeps log of past customer purchases

Data mining (knowledge discovery from data)

Web mining usually involves

Learning the application domain

Find all the interesting patterns: Completeness

Precise patterns vs. approximate patterns

Business intelligence view

Health care & medical data mining – often

Tremendous amount of data

Information integration and data warehouse construction

Frequent patterns (or frequent itemsets)

Classification and label prediction

Unsupervised learning (i.e., Class label is unknown)

#1: C4.5 (61 votes)

Database-oriented data sets and applications

Sequence, trend and evolution analysis

Are all mined knowledge interesting?

Web page analysis: from web page classification, clustering to

Efficiency and scalability of data mining algorithms

Automated vs. query-driven?

A typical kind of background knowledge: Concept

Different backgrounds/usages may require different forms

Association rule language specifications

Data mining systems, DBMS, Data warehouse systems coupling

On-line analytical mining data

Interactive mining multi-level knowledge

Integration of multiple mining functions

No coupling—flat file processing, not recommended

1989 IJCAI Workshop on Knowledge Discovery in Databases

KDD Conferences Other related conferences

Data mining and KDD (SIGKDD: CDROM)

Data mining: Discovering interesting patterns from large amounts of