Visualizing Big Data: Everything Old Is New Again
Big Data Management, Springer, Cham, (2017)DOI: 10.1007/978-3-319-45498-6_1
Abstract
Recent advances have led to increasingly more data being available, leading to the advent of Big Data. The volume of Big Data runs into petabytes of information, offering the promise of valuable insight. Visualization is key to unlocking these insights, however repeating analytical behaviors reserved for smaller data sets runs the risk of ignoring latent relationships in the data, which is at odds with the motivation for collecting Big Data. In this chapter, we focus on commonly used tools (SAS, R, Python) in aid of Big Data visualization, to drive the formulation of meaningful research questions. We present a case study of the public scanner database Dominick's Finer Foods, containing approximately 98 million observations. Using graph semiotics, we focus on visualization for decision-making and explorative analyses. We then demonstrate how to use these visualizations to formulate elementary-, intermediate- and overall-level analytical questions from the database.