Advanced visualization
-
Upload
deepu-nath -
Category
Education
-
view
545 -
download
3
description
Transcript of Advanced visualization
Advanced Visualization
Bijilash Babu Technical Architect
Technology Development Centre
NeST Software
Session on Emerging trends in
Business Intelligence
20 July 2012: Zenith Hall
Bhavani, Technopark, Trivandrum
Big Data, It’s Visualization
• Gartner’s definition of big data refers to high-volume, high-velocity and high-
variety information assets that demand cost-effective, innovative forms of
information processing for enhanced insight and decision making.
• Big Data is the convergence of three v’s: volume, variety and velocity..
• Internet of things (with different sensors), CRM, social media, etc..
• Improved use of Big Data could add t to the economy and create N jobs.
• Volume of data keeps creeping, Decision makers would struggle..
• Data visualisation would be a key for better perception.
8-Aug-12 NeST Controlled/Confidential 2
Roadmap
• Big Data
• Dimensionality
• Current trends
• Ordinary analytics
• Applied maths
• Advanced Technology
8-Aug-12 NeST Controlled 3
When big wasn’t that big
8-Aug-12 NeST Controlled/Confidential 4
• Line graph
• Stack graph
• Categories Stack graph
Track rises and falls over time
• Scatterplot
• Matrix chart
• Network Diagram
See relationships among data points
• Bar chart
• Block histogram
• Bubble chart
Compare a set of values
• Pie Chart
• Tree Map
• Analyze a text
• Word tree
• Wordle
See parts of a whole
• Mapping See the world
Timeline
8-Aug-12 NeST Controlled/Confidential 5
Source: The Economist
Better Representation
9876546765 987-654-6765
8-Aug-12 NeST Controlled/Confidential 6
Source: www.cia.gov
Better Representation
8-Aug-12 NeST Controlled/Confidential 7
Source: The New York Times
The volcano
8-Aug-12 NeST Controlled/Confidential 8
Create your own visual
8-Aug-12 NeST Controlled/Confidential 9
Source: www.wordle.net
George K. Thiruvathukal,
Associate Editor in Chief
Computing in Science & Engineering
Tag Cloud, NSF proposals
Create your own visual
8-Aug-12 NeST Controlled/Confidential 10
Created in R with wordcloud package. Data from country population. Note that the proportional sizes of China
and India were reduced in half.
Big Data
• With the exponential growth in data acquisition and generation.
• High-resolution sensors
• More disk space and more CPU cycles...
• You know, there are couple of walls around the CPU,
• and GPUs come into picture!
8-Aug-12 NeST Controlled/Confidential 11
How to go around
• Need to bring in better methods for extracting a smaller
set of relevant data
• Big Data isn’t just about numbers or volume, but the
trends – how they change over time.
• Visualisation is an invaluable tool in identifying trends
within massive data sets.
• spotting anomalies as well as outliers
8-Aug-12 NeST Controlled/Confidential 12
Calling in Maths
• Scientific Data Analysis techniques
• Numerical Linear Algebra
• SVD - The prize, compression
• PCA/ NLPCA – to reduce the dimensionality, feature extraction
Latent Semantic Indexing (LSI)
• SVM - classification, regression, and anomaly detection.
• SOM - neural network algorithm based on unsupervised learning
8-Aug-12 NeST Controlled/Confidential 13
Log plots
• Response to skewness towards large values; i.e., cases in which one or a few points
are much larger than the bulk of the data.
• To show percent change or multiplicative factors.
• Base of ten is useful when the data range over several orders of magnitude, a base
of two is useful when the data have a smaller range
8-Aug-12 NeST Controlled/Confidential 14
Better Mixing
8-Aug-12 NeST Controlled/Confidential 15
+ = ?
Visual data Mining
8-Aug-12 NeST Controlled/Confidential 16
Source: S.J. Simoff et al. (Eds.): Visual Data Mining, LNCS 4404
Advanced technology
8-Aug-12 NeST Controlled/Confidential 17
Hans Rosling...
8-Aug-12 NeST Controlled/Confidential 18
...What’s next?
Thank you!!!
8-Aug-12 NeST Controlled/Confidential 19