BITS DSECLZG555

Data Visualization and

Interpretation
BITS Pilani Dr. Reddy Rani
Pilani Campus
 BITS Pilani

Data Visualization & Interpretation

Course Introduction
Course Design

The course aims at understanding

✔ Concepts of Designing Effective Data Visualization
✔ Data Visualization using Tableau
✔ Best Practices of Dashboard Design
✔ Data Visualization using Python (Matplotlib/Seaborn/ Bokeh)

BITS Pilani, Pilani Campus

Course Learning Objectives

1. You will understand the basic principles of data visualization (DV).

You can use the learning for exploratory and explanatory analysis of
data and identify the right practices of visualization
2. You can employ best practices while designing a dashboard. You are
expected to demonstrate the same and score 70% and above in both
quizzes and assignments
3. You can construct different and appropriate visualizations, including
charts, graphs, and interactive dashboards, to efficiently communicate
insights derived from datasets of real-world data and/or client data
4. You can compare different DV methods, and effectively use the DV tool
Tableau
5. By the end of this course, you can efficiently use Python libraries
(Matplotlib, Seaborn) to generate & critique visualizations and be
able to use appropriate visualizations for data interpretation

BITS Pilani
Textbooks

Text Book(s)
T1 Data Visualisation : A Successful Design Process By Andy Kirk
T2 Storytelling with Data, A data visualization guide for business professionals, by
Cole Nussbaumer Knaflic; Wiley
T3 Information Dashboard Design: Displaying data for at-a-glance monitoring,
Stephen Few, second edition

Reference Book
R1 Matplotlib for Python Developers: Effective techniques for data visualization with
Python, by Aldrin Yim, Claire Chung and Allen Yu
R2 Show Me the Numbers: Designing Tables & Graphs to Enlighten, by Stephen
Few
R3 Visualize This: The FlowingData Guide to Design, Visualization & Statistics, By
Nathan Yau, Wiley
R4 Mastering Tableau, by David Baldwin

BITS Pilani
Course Design (cont…)

Modular Structure

M1 Data Visualizations and Practices

M2 Effective Dashboard Design
M3 Online Data Visualization Tools
M4 Data Visualization with Tableau
M5 Data Visualization with Python – 1 (Matplotlib)
M6 Data Visualization with Python – 2 (Bokeh, Seaborn)

BITS Pilani, Pilani Campus

Evaluation Components

Code Name Weight Date

Quiz1 TBD
5% [Best of 2]
Quiz2 TBD
EC1
Assignment I 12% TBD

Assignment II 13% TBD

Midsem Exam
EC2 30% Jan 18, 2025 (Saturday)
[Regular, Make-Up]

Comprehensive
EC3 Exam 40% Mar 29, 2025 (Saturday)
[Regular, Make-Up]

BITS Pilani
Course Handout

Double click to open the file

BITS Pilani
Tableau Desktop Set Up

• Students needs to obtain the licensed version of Tableau

Desktop product which is pre-requisite for hands-on sessions.
• Visit this link to register with Tableau.
• https://www.tableau.com/academic/students

• Complete the registration form.

BITS Pilani
Tableau Public

Double click to open the file

BITS Pilani
Thank You!

BITS Pilani
 SESSION 1 -PLAN

Contact List of Topic Title Text/Ref

Sessions(#) Book/External
resource
1 • Introduction T1 Ch 1
• Exploiting the Digital age
• Visualization as a Discovery tool T2 Ch 0
• Visualization skills for the masses
• The Visualization methodology
• Visualisation design objectives

12 BITS Pilani, Pilani Campus

Data Visualization is Ageless

13 BITS Pilani
Data Visualization is Ageless

Cave Art

Cave hyena - 30,000-year-old painting found in Cave hands - The art in the cave is dated between
the Chauvet Cave, France 7,300 BC and 700 AD; Argentina

14 BITS Pilani
 Classic Examples – Charles Joseph
Minardʼs map

Napoleon’s March to Moscow: 680k soldiers, to Russia in 1812, and

retreat in 1813

6 Variables in the
Data

1. The geography
2. Path taken by the army.
3. The armyʼs direction by
color (tan, and black).
4. The number of soldiers
remaining, 1mm for 10,000
men.
5. Temperature, to show the
freezing cold of the Russia.
6. Time, in relation to the
temperature, from right to
left in the bottom of the
For more explanation - https://www.youtube.com/watch?v=3T7jMcstxY0 map.

16 BITS Pilani
Classic Examples - Florence Nightingale’s Rose Chart

Depicting the avoidable deaths of soldiers during the Crimean war

4 Variables in the
Data
1. Blue area measured from the
centre to represent death from
preventable or mitigable
diseases
2. Red area, deaths from battle
wounds
3. Black – all other deaths
4. Time, deaths by each month

17 BITS Pilani
A Picture is Worth a Thousand Words

LET US DISCUSS

Scenario 1

18 BITS Pilani
Ticket trend-Showing data

19 BITS Pilani
 Ticket trend-Showing data

Ineffective

20 BITS Pilani
Ticket trend - Storytelling with data

21 BITS Pilani
Ticket trend - Storytelling with data

Effective

22 BITS Pilani
LET US DISCUSS

Scenario 2

23 BITS Pilani
Showing Popular songs

BITS Pilani, Pilani Campus

Showing Popular song

BITS Pilani, Pilani Campus

Data Visualization

What?
Graphical / Visual representation of data

Why?
⮚ Way to identify patterns, trends and outliers in data
⮚ Helps is making data-driven decisions

BITS Pilani, Pilani Campus

Purpose

“The purpose of visualization is insight, not just picture.”

Data visualization pioneer,

Ben Shneiderman

BITS Pilani, Pilani Campus

Purpose(cont..)

Visualization goals

BITS Pilani, Pilani Campus

Purpose (cont…)

Visual Processing

⮚ 30-60% of Human Brains capacity is used for visual

processing
⮚ Processes 10 million bits per second
⮚ Fastest mode of processing

BITS Pilani, Pilani Campus

Purpose (cont…)

Data Scientists

⮚ Required skill set in data driven world

⮚ Modern world needs both technical analysis and
creative storytelling
⮚ Data visualization sits exactly in middle of both
⮚ Increasingly valuable assets

BITS Pilani, Pilani Campus

Definition

“Data Visualisation is the use of Visual representation to

explore, make senesce of and communicate data”

--- Stephen Few

BITS Pilani, Pilani Campus

Data Representations

Types of Data Representations

⮚ Presentation
⮚ Visualization

BITS Pilani, Pilani Campus

Data Representations
(cont…)

Presentations

⮚ Uses data visuals to communicate

⮚ Two roles : Presenter and Audience
⮚ Communicate and Persuade

BITS Pilani, Pilani Campus

Data Representations
(cont…)

Visualizations

⮚ New term
⮚ Use visuals to think
⮚ Involves people trying to answer questions

BITS Pilani, Pilani Campus

Data Visualization Types

Based on interactivity

⮚ Static
⮚ Dynamic / Interactive

BITS Pilani, Pilani Campus

Data Visualization Types
(cont…)

Static Visualizations

⮚ Used in presentations, documents etc.

⮚ Requires careful design as its meant for offline
viewing
⮚ User can not adjust the views

BITS Pilani, Pilani Campus

Data Visualization Types
(cont…)

Interactive Visualizations

⮚ Used for exploratory data analysis

⮚ Meant for live/online interactions
⮚ Has more viewing options
⮚ User gains more control over the display

BITS Pilani, Pilani Campus

Usage of Data Visualization

Exploratory Analysis

⮚ Explore data to become familiar with data

⮚ Dig through the data
⮚ Find trends and relationships w.r.t. specific goals
⮚ Helps determine analyses to apply to data

BITS Pilani, Pilani Campus

Usage of Data Visualization (cont…)

Explanatory Analysis

⮚ Explain outcomes or results of analysis

⮚ Tell a story with data

BITS Pilani, Pilani Campus

Visualization as a discovery tool

40 BITS Pilani
Visualization as a discovery tool

• Statistical properties of the data

• Is this sufficient to understand/explore the data?

41 BITS Pilani
Visualization as a discovery tool

Graphical Presentation of the data

I. The general tendency about a

trend line in X1, Y1
II. The curvature pattern of X2, Y2
III. The strong linear pattern with
single outlier in X3, Y3
IV. The similarly strong linear pattern
with an outlier for X4, Y4

“The greatest value of a picture is when it forces us to notice what we never expected to see”
- John W Tukey

42 BITS Pilani
Visualization as a discovery tool

Ref: Same Stats, Different Graphs:

Generating Datasets with Varied
Appearance and Identical Statistics
through Simulated Annealing ACM SIGCHI
Conference on Human Factors in
Computing Systems 2017

43 BITS Pilani
References

⮚ Why do we visualize quantitative data?

https://www.perceptualedge.com/blog/?p=1897

⮚ https://www.tableau.com/learn/articles/data-visualization

⮚ https://www.tableau.com/sites/default/files/media/designing-great-
visualizations.pdf

BITS Pilani, Pilani Campus

Agenda

❑
❑ Understanding the context
❑ Effective storytelling strategies

BITS Pilani, Pilani Campus

Visualization Context

“Success in data visualization does not start with data

visualization itself.”
-- Cole Knaflic

…understanding context sets solid foundation for

data visualization creation

BITS Pilani, Pilani Campus

Visualization Context
(cont…)
Context includes :

❖ Who
To whom you are communicating?

❖ What
What do you want your audience to know or to do?

❖ How
How can you use data to help make your point?

BITS Pilani, Pilani Campus

Identifying Audience

Who?

❖ Who is audience? To whom you are

communicating?
❖ How they perceive you?
❖ Help to find common ground that helps to convey
your message

BITS Pilani, Pilani Campus

Identifying Audience
(cont…)
Who?

❖ Your audience
❑ Knowing them place you in better position for communication
❑ Be specific while identifying audience
❑ Different content for different set of audience

BITS Pilani, Pilani Campus

Identifying Audience
(cont…)
Who?

❖ You
❑ How your audience perceive you?
❑ First time interaction or established relationship?
❑ Know you as expert or need to set credibility?

BITS Pilani, Pilani Campus

Identifying the Audience-WHO

➢ You are the advertising media analyst who is made

responsible for providing recommendation on media
spend of product

51
Identifying the Audience-WHO

➢ You are the advertising media analyst who is made

responsible for providing recommendation on media
spend of product

Who : The marketing team that allocates funding for

media advertisement for a product

52
Know or Act

What?

❖ What do you want your audience to know or to

do?
❑ Action
❑ Mechanism
❑ Tone

BITS Pilani, Pilani Campus

Know or Act (cont…)

What?

❖ Action
❑ Make sure audience care about what you say
❑ You are subject matter expert
unique position to interpret the data and help lead people to
understanding and take action
❑ If no action recommendation possible / feasible, then
encourage discussion towards one

BITS Pilani, Pilani Campus

Know or Act (cont…)

What?

❖ Mechanism
❑ How will you communicate to your audience?
❑ Determines level of control and level of detail
✔ Live presentation
✔ Written document
✔ Slideument

BITS Pilani, Pilani Campus

Know or Act (cont…)

What? Mechanism

BITS Pilani, Pilani Campus

Know or Act (cont…)

What?

❖ Tone
❑ What tone do you want your communication to
set?
✔ Celebrating success?
✔ Lighting a fire to drive action?
✔ Is topic light-hearted or serious?

BITS Pilani, Pilani Campus

Identifying the Audience-WHAT

➢ You are the advertising media analyst who is made

responsible for providing recommendation on media spend of
product

Who : The marketing team that allocates funding for media

advertisement for a product

58
BITS Pilani, Pilani Campus
Identifying the Audience-WHAT

➢ You are the advertising media analyst who is made

responsible for providing recommendation on media spend of
product

Who : The marketing team that allocates funding for media

advertisement for a product

What : You want your audience (marketing team) to know that,

the current advertising campaign went well on TV but not on
print media

59
BITS Pilani, Pilani Campus
Data

How?

❖ How can you use data to help make your point?

❖ What data is available?
❑ Supporting evidence of the story

BITS Pilani, Pilani Campus

Identifying the Audience-WHAT

➢ You are the advertising media analyst who is made responsible

for providing recommendation on media spend of product

Who : The marketing team that allocates funding for media

advertisement for a product

What : You want your audience (marketing team) to know that, the
current advertising campaign went well on TV but not on print media

61
Identifying - How

➢ You are the advertising media analyst who is made responsible

for providing recommendation on media spend of product

Who : The marketing team that allocates funding for media

advertisement for a product

What : You want your audience (marketing team) to know that, the
current advertising campaign went well on TV but not on print media

How: Illustrate success with data available through analysis of

spends and product revenues

62
Context by Example

You are advertising media analyst made responsible for

providing recommendation on media spend of product

❑ Who : The marketing team that allocates funding for media

advertisement for a product

❑ What : The current advertising campaign went well on TV but find

very limited success in print media

❑ How : Illustrate success with data available

through analysis of spends and product revenues

BITS Pilani, Pilani Campus

Storytelling Strategies

Concepts for successful storytelling activity

❑ The 3-minute Story

❑ Big Idea
❑ Storyboarding

BITS Pilani, Pilani Campus

Storytelling
Strategies(cont…)

“ I would have written a shoter letter, but I did not have

the time.”
-- Blaise Pascal

…boil the overall communication to the single paragraph

or to the single, most important message

BITS Pilani, Pilani Campus

Storytelling
Strategies(cont…)
3-minute story

❑ Within 3 minutes you had to tell audience, what they need to know
❑ Removes dependence from supporting material like visuals ,
presentations etc.
❑ Need to know exactly what data is saying
❑ Being concise is more challenging than being verbose

BITS Pilani, Pilani Campus

Storytelling
Strategies(cont…)
Big Idea

❑ Boils down to a single most important sentence

❑ Per Nancy Durate (in book Resonate)
✔ It must articulate your unique point of view
✔ It must convey what’s at stake
✔ It must be complete sentence

BITS Pilani, Pilani Campus

Storytelling
Strategies(cont…)
Storyboarding

❑ Establishes structure of communication

❑ Visual outline of content
❑ Advice :
✔ Avoid using presentation software in beginning
✔ Use whiteboard, post-it notes, paper

BITS Pilani, Pilani Campus

Visualization Context

Storyboarding example

BITS Pilani, Pilani Campus

Case Study- Visualization Context
Summer Learning Programe

➢ Imagine you are a fourth grade science teacher. You just

wrapped up an experimental pilot summer learning
program on science that was aimed at giving kids exposure
to the unpopular subject. You surveyed the children at the
onset and end of the program to understand whether and
how perceptions toward science changed. You believe the
data shows a great success story. You would like to
continue to offer the summer learning program on science
going forward.

➢ Identify the Visualization Context.(WHO,WHAT and HOW)

71
BITS Pilani, Pilani Campus
Case Study- Visualization Context
Summer Learning Programe

Who : We want to communicate to is THE

BUDGET COMMITTEE, which controls the funding
we need, to continue the program.

What : Communicate the success of the program

and ask for a specific funding amount to continue to
offer it

How: Use the data collected via survey at the onset

and end of the program to illustrate the increase in
positive perceptions of science before and after the
pilot summer learning program.
72
BITS Pilani, Pilani Campus
 Case Study- Visualization Context
Summer Learning Programe

Who : We want to communicate to is THE BUDGET COMMITTEE, which controls

the funding we need, to continue the program.

What : Communicate the success of the program and ask for a specific funding
amount to continue to offer it

How: Use the data collected via survey at the

onset and end of the program to illustrate the
increase in positive perceptions of science
before and after the pilot summer learning
program.

73
BITS Pilani, Pilani Campus
 Case Study- Visualization Context
Summer Learning Programe

3‐minute story: A group of us in the science department were brainstorming about how to
resolve an ongoing issue we have with incoming fourth‐graders. It seems that when kids get
to their first science class, they come in with this attitude that it’s going to be difficult and they
aren’t going to like it. It takes a good amount of time at the beginning of the school year to get
beyond that. So we thought, what if we try to give kids exposure to science sooner? Can we
influence their perception? We piloted a learning program last summer aimed at doing just
that. We invited elementary school students and ended up with a large group of second‐ and
third‐graders. Our goal was to give them earlier exposure to science in hopes of forming
positive perception. To test whether we were successful, we surveyed the students before
and after the program. We found that, going into the program, the biggest segment of
students, 40%, felt just “OK” about science, whereas after the program, most of these shifted
into positive perceptions, with nearly 70% of total students expressing some level of interest
toward science. We feel that this demonstrates the success of the program and that we
should not only continue to offer it, but also to expand our reach with it going forward.

Big Idea: The pilot summer learning program was successful at improving students’
perceptions of science and, because of this success, we recommend continuing to offer it
going forward; please approve our budget for this program.

74
BITS Pilani, Pilani Campus
Case Study 2
BITS Pilani DSE-Electives - Discussion Topic

➢ As a faculty co-ordinator of a well-known organization, you

have to take a decision on the elective to be offered to the
students for second semester of their M.Tech Data Science
program.
➢ The available electives were
❑ Artificial Intelligence
❑ Data Visualization
❑ Systems for Data Analytics

❑ Identify the Visualization Context.

75
BITS Pilani, Pilani Campus
Case Study 2
BITS Pilani DSE-Electives - Discussion Topic

Question: Which elective should be offered for the

upcoming semester?
Data: Number of students opted for each electives.
Number of students who got placed in top companies.

76
BITS Pilani, Pilani Campus
Recap

❑ Establishing context
✔ Who
• Audience and You
✔ What
• Action, Mechanism, Tone
✔ How
• Data

❑ Story telling strategies

✔ The 3-minute story
✔ Big Idea
✔ Storyboarding

BITS Pilani, Pilani Campus

References

⮚ Knaflic, Cole. Storytelling With Data: A Data

Visualization Guide for Business Professionals,
Wiley, © 2015
❑ Chapter 1 : The importance of context

BITS Pilani, Pilani Campus

Definition Revisited

Data Visualisation is
The representation and presentation of data that
exploits our visual perception abilities in order to
amplify cognition.
--- Stephen Few

BITS Pilani, Pilani Campus

Thank You!

BITS Pilani, Pilani Campus