Visual Exploration and Analysis of Recommender Histories
Peter Hasitschka, Vedran Sabol · 2017
Content based recommender systems are commonly applied to provide automatic support to users searching for relevant information. However, as the retrieved number of resources may grow large, and because the user does not have direct control over the search process, re-finding and analyzing the retrieved information can become a difficult task. We introduce the ECHO (Explorer of Collection HistOries tool, which allows the user to visualize and analyze search result histories. Using an interactive three-dimensional scene, the history of recommender queries and the corresponding collections of recommended items can be easily explored. A multiple Levels of Detail approach empowers users to drill down starting from the overview of the query history, then performing visual meta data filtering on each result collection, down to detailed representation of single results. In particular, for each result collection we provide an interactive visualization of the meta data distribution. This representation makes it possible to apply global filters over the entire query history to identify additional valuable items in other result collections. Also, our approach supports the user in comparing the different collections and visually identifying similarities between them. ECHO is implemented as a web application. To avoid rendering performance issues and to guarantee smooth, animated transitions, we rely on graphic card acceleration using WebGL technology.