P18ITI2206-DATA VISUALIZATION
UNIT 1 – PART2
Dr. P C THIRUMAL
Department of Information Technology
Analytical Interaction , Navigation,
Techniques and Patterns
05/04/2025 DV- Unit I- Part II 2
Analytical Interactions
Information Visualization:
• Ability to clearly and accurately represent information
• Ability to interact with it to figure out what the information means
05/04/2025 DV- Unit I- Part II 3
Comparing Magnitudes
• Comparing performance of salespeople to one another
05/04/2025 DV- Unit I- Part II 4
A few typical magnitude comparisons
05/04/2025 DV- Unit I- Part II 5
05/04/2025 DV- Unit I- Part II 6
A few typical magnitude comparisons(contd..)
05/04/2025 DV- Unit I- Part II 7
05/04/2025 DV- Unit I- Part II 8
Comparing Patterns
Air Ticket Sales through Time.
Showing overall trends throughout the year and seasonal patterns
05/04/2025 DV- Unit I- Part II 9
Patterns based on the nature of data
05/04/2025 DV- Unit I- Part II 10
Strength
• Ability to see everything at once
Weakness
• Occlusion: Some bars are hidden behind others. Impossible to compare.
• Time consuming and cumbersome
05/04/2025 DV- Unit I- Part II 11
Voter Survey Question:
Are you happy with the state of the Economy
Relative height of the bar suggests that YES responses are four times greater than
NO responses, But this is not the case.
05/04/2025 DV- Unit I- Part II 12
05/04/2025 DV- Unit I- Part II 13
Difficult to see the meaningful relationship among the SORTED
values
Employee compensation
05/04/2025
per state
DV- Unit I- Part II 14
Employee Compensation & No.of Employees
California&Texas:
05/04/2025 Compensation is same.DV-
ButUnitTexas
I- Part II has more Employees 15
05/04/2025 DV- Unit I- Part II 16
Sales revenue/product Revenue & Profit per product
Sales Revenues /product
05/04/2025 DV- Unit I- Part II 17
Revenue per product sorted by product type Product type removed
Products ranked by revenue
05/04/2025 DV- Unit I- Part II 18
05/04/2025 DV- Unit I- Part II 19
Filtering(Removing)/Unfiltering(restoring)
Reducing the data that we are viewing to a subset of what’s currently there.
Purpose: To get any information we don’t need at the moment out of the way because it is distracting us from
the task at hand.
Sales 05/04/2025
of suit, coat, shoe, shirts, pants Sales of shirts and pants
DV- Unit I- Part II 20
FILTERING
Filter using Radio button Filter using Sliders
Filters the region
Two slider: One for the low end and one for the high end
Useful for filtering range of quantitative values.
05/04/2025 DV- Unit I- Part II 21
05/04/2025 DV- Unit I- Part II 22
Makes it possible to focus on a subset of data while still seeing it in context of the whole
05/04/2025 DV- Unit I- Part II 23
Highlighting
Highlighted data points in red belonging to customer in their 20s who purchased products.
05/04/2025 DV- Unit I- Part II 24
Highlighting data in one graph, resulting in the same subset of data being highlighted in other
associated graphs is called brushing or brushing and linking.
05/04/2025 DV- Unit I- Part II 25
05/04/2025 DV- Unit I- Part II 26
05/04/2025 DV- Unit I- Part II 27
Special type of aggregation
EXAMPLE:
• Country level to the state level (Drilling down)
• 05/04/2025
Individual store level to the country level ( Drilling up)
DV- Unit I- Part II 28
Re-Expressing
Change in unit of Measurement :USD to %
05/04/2025 DV- Unit I- Part II 29
Monthly Sales Revenue change through Time Each month’s sale compared to January
05/04/2025 DV- Unit I- Part II 30
05/04/2025 DV- Unit I- Part II 31
05/04/2025 DV- Unit I- Part II 32
Re-Visualizing
• Switching from one type of graph to another quickly
and easily.
• A GRAPHIC is no longer ‘Drawn’ once and for all: It is
constructed and reconstructed until all the relationship
which lie within it have been perceived.
• A Graphic is never an end itself: It is a moment in the
process of decision making.
• No single way of visualizing data can serve every
analytical needs.
• Different types of visualization have different
strengths.
05/04/2025 DV- Unit I- Part II 33
Re-Visualizing
Variation between actual and budgeted expenses change
through the year
05/04/2025 DV- Unit I- Part II 34
05/04/2025 DV- Unit I- Part II 35
Zooming and Panning
Bottom05/04/2025
Graph: Zooming of data from Feb-14 to Feb-20 (Specific period)
DV- Unit I- Part II 36
Panning: View areas of the graph that reside outside the boundaries of the current
magnified position.
We can05/04/2025
change the view by moving up, down, left or right
DV- Unit I- Part II 37
05/04/2025 DV- Unit I- Part II 38
1. Linear Scale
2. Logarithmic scale
Which is increasing at faster
Slopes the two lines are identical.
rate (hardware or software)
Hence Both are increasing at the
Ans: Hardware
same 10% rate of change.
Linear : Distance between tick marks are equal
Logarithmic: Log scale with base value 10.Useful when we compare rate of change
05/04/2025 DV- Unit I- Part II 39
05/04/2025 DV- Unit I- Part II 40
Details on Demand
•Pop-up box (Tool Tip) containing details appears as we hover with mouse over a particular item in a graph.
•It disappears when we move the mouse
05/04/2025 DV- Unit I- Part II 41
05/04/2025 DV- Unit I- Part II 42
Annotating-Make notes
05/04/2025 DV- Unit I- Part II 43
When notes are associated with particular items in a visualization and those items change
position, the notes should automatically reposition to maintain the association.
05/04/2025 DV- Unit I- Part II 44
05/04/2025 DV- Unit I- Part II 45
Analytical Navigation
05/04/2025 DV- Unit I- Part II 46
Directed: Begins with a question and searches for an answer and then produce the answer
Exploratory: Begins by simply looking at the data, notice something interesting and ask a
question
05/04/2025 DV- Unit I- Part II 47
Overview
05/04/2025 DV- Unit I- Part II 48
Zoom
05/04/2025 DV- Unit I- Part II 49
Filter and Details on Demand
05/04/2025 DV- Unit I- Part II 50
Node-Link Visualization
Tree Diagram
representing hierarchical
data
05/04/2025 DV- Unit I- Part II 51
Tree Map
05/04/2025 DV- Unit I- Part II 52
Tree map
• The treemap chart is used for representing
hierarchical data in a tree-like structure. Data,
organized as branches and sub-branches, is
represented using rectangles, the dimensions and
plot colors of which are calculated w.r.t the
quantitative variables associated with each
rectangle—each rectangle represents two numerical
values. You can drill down within the data to,
theoretically, an unlimited number of levels. This
makes the at-a-glance distinguishing between
categories and data values easy.
05/04/2025 DV- Unit I- Part II 53
05/04/2025 DV- Unit I- Part II 54
General techniques and practices that can improve the effectiveness of visual
analysis
05/04/2025 DV- Unit I- Part II 55
05/04/2025 DV- Unit I- Part II 56
Bar Graph-Begin the Scale at Zero
Expense of Sales is 4.5 greater than expense of
marketing ( Our eyes perceive difference)
05/04/2025 DV- Unit I- Part II 57
05/04/2025 DV- Unit I- Part II 58
05/04/2025 DV- Unit I- Part II 59
Reference Lines
Top: Mean sales of products in each region
Bottom: 05/04/2025
Average sales revenues for products in all regions(Reference
DV- Unit I- Part II line is the same in each graph) 60
05/04/2025 DV- Unit I- Part II 61
Relative ease. It is called as Trellis display/small multiples
05/04/2025 DV- Unit I- Part II 62
Trellis Displays - Horizontal
05/04/2025 DV- Unit I- Part II 63
Trellis Displays - Vertical
05/04/2025 DV- Unit I- Part II 64
Trellis Displays–Matrix(Ordered on
Department)
05/04/2025 DV- Unit I- Part II 65
Trellis Displays–Matrix(Ordered-Expenses)
Trellis : Graphs differ by one variable
05/04/2025 - Department
DV- Unit I- Part II 66
Visual Crosstab
Arrange
05/04/2025
graphs differ by moreDV-than one variable
Unit I- Part II 67
Crosstabs
05/04/2025 DV- Unit I- Part II 68
05/04/2025 DV- Unit I- Part II 69
05/04/2025 DV- Unit I- Part II 70
Blind men: Two front legs- Two big tree without branch, Tail- straw fan swinging back and
forth to give us breeze, trunk – snake . The experience of each man was unique because each
experienced different part of the elephant and that part alone.
When we can’t examine data from multiple perspective simultaneously, many of the meaningful relationships
that exist in our data will remain hidden
05/04/2025 DV- Unit I- Part II 71
Multiple Concurrent Views
05/04/2025 DV- Unit I- Part II 72
Multiple Concurrent Views contd..
05/04/2025 DV- Unit I- Part II 73
05/04/2025 DV- Unit I- Part II 74
05/04/2025 DV- Unit I- Part II 75
05/04/2025 DV- Unit I- Part II 76
05/04/2025 DV- Unit I- Part II 77
In graphs that use data points or lines, multiple objects sharing the same space,
positioned on top of one another. Makes
05/04/2025 it difficult
DV- Unit I- Part II to see individual values. 78
Reducing the size of the data object
05/04/2025 DV- Unit I- Part II 79
Over –Plotting Reduction
Removing fill color
05/04/2025
Changing the shape
DV- Unit I- Part II 80
Over –Plotting Reduction
Jittering Data Objects: Altering the actual values. Moving them
to slightly different positions.
Making the data object Transparent: Proper degree of
transparency allows us to see through the objects to perceive
difference in the amount of over-plotting as variations in color
intensity.
Encoding the Density of Values:
• Contour lines to outline areas that contain varying densities
of data points.
• Subdividing the scatter plot into small regions and display
the number of data points in each region as colors of varying
intensities.
05/04/2025 DV- Unit I- Part II 81
05/04/2025 DV- Unit I- Part II 82
05/04/2025 DV- Unit I- Part II 83
Thank You
05/04/2025 DV- Unit I- Part II 84
05/04/2025 DV- Unit I- Part II 86
05/04/2025 DV- Unit I- Part II 87
What is information? According to Stephen Few, an educator and innovator in the field of information
visualization, information comes from items, entities, and things that cannot and do not have a direct
correspondence to physical structures or objects.6 Some good examples include football statistics, stock
market prices, connections between socioeconomic status and criminal rates, and relationships
between car attributes and mileage per gallon. On the other hand, examples for an entity that has a
correspondence to physical structures or objects include human anatomy and three-dimensional cell
structure. The “information” in this context is abstract, as it comes from an analysis of some type of
data. The second part of information visualization is visualization. Visualization refers to the creation of
two-dimensional or three dimensional representations of data that enable new discoveries of both
insights and knowledge. With the close connection between human vision and cognitive capacity,
visualization can also be seen as the use of computer-supported, interactive visual representations of
data to enhance cognition. Together, these two words describe a new meaning that has changed the
way we perceive information and understand data in a highly impactful, more memorable manner
05/04/2025 DV- Unit I- Part II 88
05/04/2025 DV- Unit I- Part II 89
05/04/2025 DV- Unit I- Part II 90
05/04/2025 DV- Unit I- Part II 91
Visualization- Goals
05/04/2025 DV- Unit I- Part II 92
Purpose of Information
Visualization
• According to Smashing Magazine cofounder Vitaly Friedman:
“ability to visualize data, communicating information clearly and effectively” but
also “in a more intuitive way.”
The Institute of Development Studies:
A way “to explain and to explore data” to “be used as a tool for analysis, finding
patterns as well as discovering questions amongst other things.”
Today, the way we define information visualization is grounded in the visual
elements, and in particular pictorial or graphical formats.
Its key use has been identified as its ability to help decision makers see analytics,
further helping them to comprehend difficult concepts and even identify new
patterns. The evolution of digital technologies has only broadened the use of
information visualization as it is now being used to extract information from data
for more detail. Current trends and demands show us that information
visualization is now interactively changing the way the human brain visualizes
and processes complex data. Information visualization is easier for the brain to
process than other forms of data, such as reports or spreadsheets.
05/04/2025 DV- Unit I- Part II 93
05/04/2025 DV- Unit I- Part II 94
05/04/2025 DV- Unit I- Part II 95
05/04/2025 DV- Unit I- Part II 96
05/04/2025 DV- Unit I- Part II 97
05/04/2025 DV- Unit I- Part II 98
