Visualization Techniques for Extracting Information from Large Data Volumes

Abstract

According to Google CEO Eric Schmidt in 2010 the humanity was producing in 2 days the amount of recorded data that it would produce from the dawn of the civilization up until year 2003 [1]. These enormous amounts of data can be used to extract useful information in various fields of science, medicine, security, politics, and community activities. In order to do this, however, expertise in the subject domain is necessary. Unfortunately, the professionals in specific areas, especially the non-technical ones such as humanities, arts, politics, and sport do not have the skills and knowledge to write elaborate computer programs that would process their data. On the other hand, if they were able to present the data visually in an appropriate way, they could use their intuition and draw conclusions and make decisions. In this work we give an overview of some of the most common free visualization techniques for extracting information from large data volumes. Our goal is to unveil their potential so that non-expert could get an idea how they could empower their work. More specifically, we will focus in our work on Google Analytics, Pentaho, and R. Google Analytics is a data visualization tool intended for website metrics, such as daily access, unique users, time spent on the website, and so on. Pentaho is a tool for Business Intelligence that provides tools from basic reports to predictive modeling. R is language and environment for statistical computing and graphics, providing a variety of techniques. It is open source software and highly expendable. We explain and compare the above software systems from non-technical perspective and consider several case studies. We will also discuss the reasons of using visualization tools on a real-life example of raw data and its visualization with the right tool. Our presentation is intended to a broad audience in no specific technological knowledge. References: [1] Siegler, M.G., Eric Schmidt: Every 2 Days We Create As Much Information As We Did Up To 2003. TechCrunch, Aug. 4, 2010.