Data Visualization

Angela Lu

Week 1
Information Systems, CB
City University of Hong Kong
About me…
• Assistant Professor of Information Systems
• PhD in sociology from Stanford University
• Research areas: sharing economy, IT career, IT innovation,
computational social science, social network analysis

• Prior background in data visualization

• Designed and taught the undergraduate version of this course
(IS4335)

2
What is data visualization?
• Definition: computer-based visualization systems provide
visual representations of datasets designed to help people
carry out tasks more effectively

• Visualization is suitable when there is a need to augment

human capabilities rather than replace people with
computational decision-making methods

3
Why visualize data?
• Data visualization helps decision making
• Big data is messy and hard to make sense of
• A graph is worth 1000 words (intuitive to understand)
• Present details in data
• Helps formulate interesting questions
• Speed up analytical process

4
Intended goals
• Know ways to visualize and present data, using software
packages such as R, Tableau, and Python
• Know what design is effective, and why
• Extract information from visual presentations; know what
questions to ask

5
Intended audience
• No prerequisites
• No assumed coding background
• Many areas helpful but not required
• computer programming, statistics, business analytics

• Open to master’s students who want to pursue business

analytics and information systems as a career

6
Schedule
• Lectures on Wednesday (S61) and Thursday (S62), 7-9:50 pm
• Semester A 2022/23, 13 weeks
• Materials are posted on Canvas
• Data visualization tools: R, Tableau, Python
• 50 min tutorial, 50 min lab exercises
• No tutorial during first week, only lab organizations and group
sign-up

7
Modules
Basics Data structure Advanced Topics

Design analysis Tables (w4-5)

framework Computer
(w2) vision (w10)
Spatial data
(w6) Intro to deep
Marks and learning
channels (w3)
(w11)
Networks
(w8-9)
8
Assignments & grading
• Individual project, 40%
• Lab exercises, 30%
• Submit exercises from week 3 to week 9
• Group project, 20%
• Class participation, 10%

9
Individual project (40%)
• Similar to a research paper
• Completed individually
• Apply visualization techniques and concepts to address a real-
world issue
• Research question description
• Data description
• Design visuals and interpret insights
• Research and business implications
• Comprehensive written report (5-10 pp.)

10
Lab exercise (30%)
• 50 min tutorial, 50 min in-class exercises, 10 min break in
between
• Weekly exercises in R and Tableau (w3-w9), complete individually
and submit on canvas
• 10-15 min of discussion of the answers at the beginning of next
tutorial
• Get full credit as long as you submit at the end of the week
• Late submission is only allowed in exceptional circumstances

11
Group project (20%)
• Group of 5-8 students (self sign-up on Canvas)
• Apply visualization tools to form an analytics report
• Identify a data source. Resources on datasets later on
• Find an interesting problem about the data
• Use R, Python or Tableau for analysis and visualization
• Project milestone report submitted in week 6: 2-3pp on data
source, analytical steps, and initial findings
• Project milestone presentation in week 7: 5 minutes each group
• Final presentation: 15 min including Q&A, in the last 2 weeks

12
Textbooks
• Visualization Analysis and Design (VAD). Tamara Munzner, 2014
• Lectures will follow this

• R and Tableau references:

• R: Data Analysis and Visualization. Tony Fischetti and Brett Lantz, 2016
• Tableau Your Data!: Fast and Easy Visual Analysis with Tableau
Software. Daniel G. Murray, 2016

• Python reference:
• François Chollet, 2018. Deep Learning with Python.
13
Software resources
• R:
• Download: https://www.r-project.org/
• Rstudio download: https://www.rstudio.com/products/rstudio/download/

• Tableau Public
• https://public-pantheon.tableau.com/en-us/s/download

• Python:
• Python: pre-installed in mac OS
• Anaconda distribution: https://www.anaconda.com/products/distribution

14
Today’s outline
• What is data visualization, and why use it
• Examples

15
What is data visualization?
• Definition: computer-based visualization systems provide
visual representations of datasets designed to help people
carry out tasks more effectively
• Visualization is suitable when there is a need to augment human
capabilities
• It does not replace human decision making, but makes it better
• Visualization examples

16
17
18
19
Why visualize data?
• The advantage of visualization
• Large datasets are hard to represent by hand
• Visualization is intuitive; more effective representation
• Present details otherwise hard to see
• Dynamic trends over time

20
Gene interaction: messy big data
21
Computer-based visualization systems provide visual representations of datasets
designed to help people carry out tasks more effectively.

• summaries lose information, details matter

confirm expected
–Summaries and find unexpected patterns
lose information;
–details
assessmatter
validity of statistical model
[Anscombe, 1973]
Anscombe’s Quartet
Identical statistics
x mean 9
x variance 10
y mean 7.5
y variance 3.75
x/y correlation 0.816
https://www.youtube.com/watch?v=DbJyPELmhJc

Same Stats, Different Graphs 6

Why show details? Statistical properties can be misleading…

22
23
What design is better?
• Main concern: effectiveness of visualization design for
particular tasks

• Why some designs are more effective than others?

24
Search space metaphor for vis design
25
Which design is better?

26
 A: Information density is low

C: Good tradeoff btw density and spatial position

B: Information is compact; spatial position hard to read
Which design is better?

27
What design is better?
• Main concern: effectiveness of visualization design for
particular tasks

• Why some designs are more effective than others?

• Is there a way to measure design effectiveness? Introducing

an analytical framework next time…

28
Next …
• Lab organization

29
Lab Organization
• Provide hands-on experience of data visualization
• Try concepts and materials covered in lectures
• Use software packages, such as R, Tableau, and Python

• Lab components…

30
Lab Components
• 50 min tutorial
• 50 min exercises
• Submit exercises on canvas before Sunday midnight the
same week
• Group discussion of exercises is encouraged
• No tutorial this week, only demos and group sign-up

31
Lab exercises
• Small exercises in R and Tableau (w3-w9)
• Get full credit as long as you submit before the deadline, typically
Sunday the same week
• Late submission is only allowed in exceptional circumstances;
email us in advance

32
Visualization software for the lab
• Tableau Public: https://www.tableau.com/products/public/download
• R: https://www.r-project.org/
• Rstudio: https://www.rstudio.com/products/rstudio/download/
• Python Software
• Python: pre-installed in mac OS
• Anaconda distribution: https://www.anaconda.com/products/distribution

• Today:
• Software demos
• Software installation
• Group sign-up

33
What is Tableau?
• Tableau: one of the fastest growing business analytics and data
visualization tools; easy to share analytics in the cloud
• Good for fast analytics
• Big data; built-in calculations
• Live data update and connection
• Little programming required; drag and drop visualization

• A short demo …
• https://www.youtube.com/watch?v=YfE9jBq002s&ab_channel=T
ableauSoftware
34
Tableau Public
• Tableau public installed in lab, free version, great for sharing
• Download: https://www.tableau.com/products/public/download
• Features of Tableau public…
• Connectivity: Can connect to excel and txt files; cannot connect to
databases
• Security: Cannot save your workbook in private
• Anything you save is saved on Tableau public server (no confidentiality)
• Data limit: limited to 10 mil rows in a single connection
• Account space: 10 GB
• Check out cool viz examples:
• https://public.tableau.com/en-gb/gallery/?tab=viz-of-the-
day&type=viz-of-the-day

35
 38

R visualization
• Data table
• Spatial & map
• network
What is R?
• R is a statistical programming language for data analysis
• Powerful, free, similar to Matlab
• You can do almost anything in R, besides data visualization
• Programming required
• Popular IDE (integrated development environment): RStudio

39
R download and installation
• Download R first
• R for Windows:
• https://cran.r-project.org/bin/windows/base/
• R for Mac OS:
• https://cran.r-project.org/bin/macosx/

• Download Rstudio:
• https://www.rstudio.com/products/rstudio/download/
40
Ggplot2 examples

41
 42

Python
Visualization
Python Installation
• Python: pre-installed in Mac OS
• Python for Windows:
• https://www.python.org/downloads/windows/
• Anaconda distribution:
https://www.anaconda.com/products/distribution

• Basic libraries: numpy, matplotlib, ggplot, plotly, seaborn …

43
Installing Jupyter Notebook: Easy Way
• Prerequisite: Python
• While Jupyter runs code in many programming
languages, Python is a requirement (Python 3.3 or greater, or
Python 2.7) for installing the Jupyter Notebook
• We recommend using the Anaconda distribution to install Python
and Jupyter
• Download Anaconda:
https://www.anaconda.com/products/distribution
• See example notebook on Canvas
• Youtube demo: https://www.youtube.com/watch?v=a9UrKTVEeZA

44
Alternative: Pip Install
• Open terminal (if using Mac or Linux), or command line in
Windows
• First install pip
• curl https://bootstrap.pypa.io/get-pip.py -o get-pip.py
• python get-pip.py
• Then install Python & Jupyter notebook
• pip install python
• pip install jupyterlab
• pip install notebook
• Open Jupyter Notebook
• jupyter notebook

45
46
47
Group Sign-up
• Form groups of 5-8 students for group project
• Sign up on Canvas

48
Questions?
• Contact:
• Angela Lu: qinglilu@cityu.edu.hk
• Jingwen Pan (TA): jingwepan4-c@my.cityu.edu.hk
• Subject line: include “IS6335”

• Office hours: Monday 3-4pm, Lau 6-270

49