Scribd
Upload
17 views66 pages

AI For Everyone Unit V

Uploaded by

intern
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
17 views66 pages

AI For Everyone Unit V

Uploaded by

intern
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 66

Data Visualization

DATA VISUALIZATION IS
COMMUNICATION
Principles of Graphical Excellence
• Clarity

• Precision

• Efficiency

• Maximize ideas, minimize ink


Data Viz in a Nutshell

1. History

2. Graphical Basics
3. Details to Make Your Viz More
Understandable
Data Viz in the 10 Century
th
Maps as Rich Information Source
Spatial Patterns Intuitive
Sometimes One Variable is Enough
Adding Space to Time
Unifying Maps and Timeseries
Abstraction Can Tell Wonderful Stories
Data Viz in a Nutshell

1. History

2. Graphical Basics
3. Details to Make Your Viz More
Understandable
Elements of a Plot

Geometric Objects Scales & Coordinates Annotations


Elements of a Plot
Aesthetics of a Plot
Aesthetics Map Data to Visual
Representation
Aesthetics Map Data to Visual
Representation
Aesthetics Map Data to Visual
Representation
Types of Visualizations: Amounts
Types of Visualizations: Amounts
Barplots
What is Wrong Here?
What is Wrong Here?
Bars are Not Always Right
Types of Visualizations: Distributions
Histograms Show Frequency
Types of Visualizations: Multiple
Distributions
Boxplots to Show Variation
…Or Point-Ranges
Visualizing Uncertainty is Difficult
What is Wrong Here if Bars Represent 2SD
of Temperature?
Visualizing Distributions versus Uncertainty
How Does this View Differently?
It’s Tempting to Plot Raw Data, but, Be
Careful…
Lots of Ways to Visualize Data,
Distribution, and Uncertainty

https://dallasnova.rbind.io/post/efficient-data-visualization-with-faded-raincloud-plots-delete-boxplot/
From rweekly.org 9/5 edition https://rweekly.org/2022-W36.html
Relationships Between Variables
Relationships Between Variables With a Lot
of Data
Scatterplots Show Relationships
Correlations to Understand Relationships
Alternate Ways to View Correlations
Efficiently

https://albert-rapp.de/posts/ggplot2-tips/13_alternative_corrplots/
13_alternative_corrplots.html
Should Your Data Be Connected?
Geospatial Visualization
Combining Data Sources and Maps

• Chlorophyll a
• Abundance of taxonomic groups
• Temperature
• Salinity

Hopkinson and Hobbie 2012 http://ecosystems.mbl.edu/PIE/data/EST/EST-PR-ChemTax.html


Visualizing Uncertainty in Relationships is
Difficult
Data Viz in a Nutshell

1. History

2. Graphical Basics
3. Details to Make Your Viz More
Understandable
Coordinate Systems Transform
Relationships
The Polar Coordinate System is Useful!
The Polar Coordinate System is Useful!
Adding Full Scale to 0
Why Transform?
Log-Transformation To See Relationship
Scales to Add Dimensions of Data
Colors Can Distinguish Groups
Colors Can Show Data
Beware Not Thinking About Color
Blindness

Never use Red-Green!


Use redundant coding: shapes, sizes, etc.
Redundant Coding
Colorbrewer.org

See also https://flowingdata.com/tag/color/


A Touch of Color!
Color Can Bring in Another Dimension
Statistical Fit to Aid Understanding
Facets to Add Fine-Grained Information or
New Dimensions
Facets Add Information
Finally: Minimalist Presentation
1. Above all else show data.

2. Maximize the data-ink ratio

3. Erase non-data-ink.

4. Erase redundant data-ink.

You might also like