Experimental Evaluation of a Stacked Graph Visualization
Alejandro Szanto de Toledo, Ruck Thawonmas, Frank Rinaldo · 2010
We introduce an experimental evaluation of a Stacked Graph Visualization ( SGV) that was designed to facilitate the production of views required for different visual analytics tasks. Our interface encourages users to gain insight into a data set of 510 time series of 255 kinds of jobs. We include a case stud y that applies the SGV in a visual analytics environment and a usability study that evaluates the correlation be tween three implicit interest indicators (dwell time, view revisits, and mouse activity) and the explicit rating for a single view. The SGV was developed to record the views produced by 14 users and the explicit rating of those vie ws. Using the data collected by the SGV, the individual ratings and some combinations of explicit ratings were analyz ed and compared with the explicit rating. We found that the three implicit interest indicators had a strong corre lation with explicit interest.