Virtual Reality in Visual Analytics of Large Datasets

Introduction

In the digital age we are living in, huge data is generated every second. So the information overload is a common phenomenon. To deal with this some efficient computation techniques, storage devices, as well as analytical techniques, are desired. Analysis of such unstructured data, which are complex in nature and are huge in volume, is tedious. Analysts usually take help of some analysis tools. These tools can help to extract significant information from the ocean of raw data[Andrew Moran; Vijay Gadepally; Matthew Hubbell; Jeremy Kepner,2015]. Some tools used for visual analytics will be discussed briefly later in the chapter. Conventional ways to represent data were not sufficient to represent big-data. Visual analytics serve the purpose to fill the gap between conventional methods of data representation and those methods which are needed for big data with the added intelligence. Three major goals of visualization are (a) presentation (b) confirmatory analysis (c) exploratory analysis[Keim D.A., Mansmann F., Schneidewind J., Thomas J., Ziegler H., 2008]. Presentation refers to the presenting of the data to the user in a manner that is easily understandable and further processed whereas confirmatory and exploratory analysis refers to the process in which whether the data after processing is correct and usable or not is checked and further prospective decisions are taken.

Since big data is represented by a characteristic of being voluminous, it presents the unquestionable trial of data representation and visualization. Existing methods need to improvise with the ever-changing requirements of the system. Conventional methods for visual analytics use graphs, charts, trees, gauges etc. Virtual Reality if induced with VA can create a powerful environment for the analyst. VR provides an interactive and immersive environment for the user. Due to the same, it makes information easy to perceive and easy to process. Various big industries like IBM, Microsoft, Google etc offers solutions for big data handling. Such visualization tools are helpful in business as well as research perspectives. As data processing and generating speed is increasing exponentially, the need for replacing conventional analytical methods increases too. This paper provides detailed discussion over data representation, an analysis of various existing tools, induction of Virtual Reality in the VA environment and further research possibilities in the same.