Download - Data visualization
Data Visualization
@rohitnotifies
The goal of data visualization is to tell a story, without telling
lies.
Data analysis is a cognitive problem.
– We are building visualizations to assist humans with cognitive
processes
– Helping to make decisions
– Influence decisions
– Ethically
It is important that you get to the point where the problem
drives your choice of visualization intuitively.
Data Visualization
@rohitnotifies
Choropleth, or filled maps, are the most common.
Use a color gradient to denote the quantitative element of
interest
Choropleth maps can only represent the level of a single
measure.
Visualizing Maps
@rohitnotifies
A dashboard is a visual display
of
the most important information needed to achieve one or more
objectives
that has been
consolidated on a single computer screen
so it can be
monitored at a glance
Stephen Few (Information Dashboard Design, p 26)
Dashboard
@rohitnotifies
They should communicate something
They should fit entirely on one screen
• No scrolling
• No pages
They are an overview.
They should quickly point out when something is wrong.
Dashboard - Specific Objective
@rohitnotifies
1. Remember that a dashboard must be fit for purpose.
• Right information
• Right audience
• Right design
2. Know your dashboard type
• Strategic/Executive
• Analytical
• Operational
When building a dashboard, the relationships between the
various elements must be fully explored.
When building a dashboard, you need to be aware of its
affordances.
Dashboard - Best Practices
@rohitnotifies
Text analytics are a huge focus in the world of big data
• Twitter, Facebook, etc.
• Documents
• Blogs
• Comments on blogs
Of course, you can visualize the results of your text mining using our
normal set of visualizations.
The problem with all of these is that the terms are removed from context.
Text visualization has lagged behind unstructured data visualization
Text Visualization
@rohitnotifies
Simple text visualizations say a great deal
• Word/Tag Clouds
• Word Trees
• Term Networks
There are packages for R and many free tools that can generate these
visualizations.
Text Visualization
@rohitnotifies
Word/Tag clouds (First known appearance in 1992, in Germany)
• Approximate the ratio of word count
• Are computationally heavy (but getting better)
• Help to visualize important terms
• Can be easily cluttered.
• Nothing to direct attention besides size
• This means that the human eye focuses only on the largest words.
• Are most of the words even looked at?
Text Visualization
@rohitnotifies
New – most techniques have been developed in the past 4 years
http://textexture.com/index.php?text_id=604
Text Network Visualization
@rohitnotifies
Speechopedia - Ex - Prime Minister, Dr. Manmohan Singh
https://gramener.com/speechopedia/
@rohitnotifies
Speechopedia - Ex - Prime Minister, Dr. Manmohan Singh
https://gramener.com/speechopedia/
@rohitnotifies
Show the data
Induce thinking about the substance
Avoid distorting what the data has to say
Present many numbers in a small space
Make large datasets coherent
Encourage the eye to make comparisons
Reveal data at several levels of detail
Excellent Graphics
@rohitnotifies
Data becomes evidence when the relationships between
variables are illustrated.
Chart Junk - Style over substance
Chartmakers reveal what they choose to, and can mislead
audiences.
Induce thinking about the substance
“Above all else show the data.” Tufte
Takeaways
@rohitnotifies
12 data maps that sum up London - BBC
Busiest rail stations
Busiest rail stations
Contact Details
+91- 9989 071 326
@rohitnotifiesRohit Kumar Cherukuri
www.DigiMaverick.com
C Rohit Kumar