Location: Department of Computer Science, University of Hong Kong
Exploratory Data Analysis (EDA) is an iterative and tedious process. Several strategies have been proposed to ease the burden on users in EDA ranging from stepwise to full-guidance approaches. Stepwise approaches rely on computing utility functions that determine the best action to take at each step. Full-guidance approaches rely on learning end-to-end exploration policies. Today’s big question is how to commodify EDA and make it easily deployable for all but for that we need to know what users are looking for: are they looking for a needle in a haystack, taking a tour of the data, or are they feeling lucky? This talk will investigate those questions and discuss the challenges of storing learned pathways through data or regenerating them when needed.