Title: Using Bayesian visual analytics to conceptualize uncertainty and explore data
Authors: Leanna House - Virginia Tech (United States) [presenting]
Abstract: While inundated with big data, uncertainty is also present. There is uncertainty in data collected, uncertainty in methods used to summarize data, and uncertainty in judgements formed from data summaries. Alas, in the presence of big data, data analysts often avoid quantifying uncertainty formally and/or avoid communicating degrees of uncertainty in what might be gleaned from visual or quantitative summaries. There are many potential reasons for avoiding uncertainty, ranging from difficulty (it is hard to model big data well) to misplaced confidence in laws of large numbers to inconsistencies in human behavior. Is it well accepted that, even when measured by probability, the ways by which humans interpret, process, and use uncertainty are personal and vary widely. We take a visual analytic and probabilistic approach to engage humans in learning from big data visually, while considering uncertainty. Specifically, we start with a method we developed called Bayesian Visual Analytics (BaVA) and incorporate novel, visual metaphors of uncertainty, in the context of weighted multi-dimensional scaling visualizations. The door is open to future research into how humans incorporate uncertainty as they visually explore and learn from big data.