Insight in to Data Visualization

Data visualization is a method of displaying data in more understandable form which makes it easy to interpret, associate and analyze in comparatively shorter duration of time. It is not mere presentation of data through art to make it aesthetically pleasing; it involves a set of well-defined rules based on science of visual perception & cognition that must be followed for effective display of information and for quick understanding. It includes transformation of numeric data into visual form.

Data visualization is done to understand the hidden facts, trends, patterns and relationships in the data [1]

Why Data visualization
Visual science has proved that data visualization can communicate and explain the concepts more efficiently, efficiently and in shortest time. If visualized correctly it increases both speed and retention of data (knowledge). Table below shows the benefits of data visualization according to the respondent percentages of a survey. These are the proven results taken from published source [2]

Before we dig into the techniques of data visualization and the tools available for the same, it is important to mention that Data Visualization Applications scope is tremendous! According to Mordor Intelligence [3] the data visualization market will increase at a compound annual growth rate (CAGR) of 9.21 % from $4.12 billion in 2014 to $6.40 billion by the end of 2019!

Methods and techniques for Data Visualization:
Presently data analysis techniques are based on tools of statistics and computer science. The advance techniques to deal with the large sized data are Natural Language Processing, Biological Neural Networks, Artificial Neural Networks, Statistics, and Methods of Predictive Analysis etc. In most cases these techniques are used simultaneously for data analysis, and processing using them interconnectedly improves system's effectiveness.

Some of the methods and techniques of data visualization are:
Tag cloud: visual depiction (font size, color etc) of online content to represent the prominence or frequency to speed up the processing.
Clustergram: shows the relation between each element present in the data whenever there is change the data. It is basically imaging technique used in cluster analysis.
Motion charts by Goggle are dynamic bubble chart for visualization of multivariate Data. amCharts a company based out of Lithuania offers JavaScript charts, maps, stock charts for data visualization.

Software and Tools for Data Visualization
Pentaho: It supports the spectrum of Business Intelligence (BI) functions such as analysis, dashboard, enterprise-class reporting, and data mining.
Flare: An Action Script library for creating data visualization that runs in Adobe Flash Player.
JasperReports: It has a novel software layer for generating reports from the big data storages.
Dygraphs: It is quick and elastic open source JavaScript charting collection that helps discover and understand opaque data sets.
Datameer Analytics Solution and Cloudera: Datameer and Cloudera have partnered to make it easier and faster to put Hadoop into production and help users to leverage the power of Hadoop.
Platfora: Platfora converts raw big data in Hadoop into interactive data processing engine. It has modular functionality of in-memory data engine.
ManyEyes: It is a visualization tool launched by IBM. Many Eyes is a public website where users can upload data and create interactive visualization.
Tableau: It is a Business Intelligence (BI) software tool that supports interactive and visual analysis of data. It has an in-memory data engine to accelerate visualization.

Technical Challenges Pertaining to Data Visualization:
There are following challenges pertaining to data visualization implementation especially in Big Data. Separation of objects on screen is not easy if they are too related; commonly refer to as Visual Noise. Information loss - when reducing data sets into visible objects there could be a problem of information loss. Also data visualization methods are lemmatized by physical perception limits along with limitation of aspect ratio and resolution of the device. Other challenges are high rate of image change, high performance requirements etc.

By Ms. Poonam - Assistant Professor, CSE, Chitkara University H.P.

References:

• IDRC|CRDI (International development and research centre Canada)"Data Visualization in Review: Summary "Written by Jacqueline Strecker, the Evaluation Unit's research awardee for 2011-2012.
• V. Sucharitha, S.R. Subash and P. Prakash , Visualization of Big Data: Its Tools and Challenges, International Journal of Applied Engineering Research, 9(18), 2014, pp. 5277-5290.
• Report: Data Visualization Applications Market Future Of Decision Making Trends, Forecasts And The Challengers (2014 - 2019). Mordor Intelligence; 2014.
• Wang, Lidong, Guanghui Wang, and Cheryl Ann Alexander. "Big Data and Visualization: Methods, Challenges and Technology Progress." Digital Technologies 1.1 (2015): 33-38.
• "Data Visualization Past, Present, And Future "by Stephen few perceptual edge Wednesday, January 10, 2007 in cognos.