Publication Date:
2014-08-14
Description:
Taste in music is of highly subjective nature, making the recommending of music tracks a challenging research task. With TRECS, our live prototype system, we present a weighted hybrid recommender approach that amalgamates three diverse recommender techniques into one comprehensive score. Moreover, our prototype system peppers the generated result list by some simple serendipity heuristic. This way, users can benefit from recommendations aligned with their current taste in music while gaining some exploratory diversification. Empirical evaluations of the live TRECS system, based on an online evaluation, assess the overall recommendation quality as well as the impact of each of the three sub-recommenders. In addition, to better understand the nature and impact of serendipity in isolation, we conducted another study with another recommender prototype of ours, named SONG STUMBLER. The latter assesses three different serendipity metrics in an online evaluation. Content Type Journal Article Pages 235-248 DOI 10.3233/WIA-140294 Authors Cai-Nicolas Ziegler, XING EVENTS GmbH, Sandstraße 33, D-80335 München, Germany Thomas Hornung, REWE Information Systems GmbH, Humboldtstraße 140-144, D-51149 Köln, Germany. E-mail: thomas.hornung@rewe-group.com Martin Przyjaciel-Zablocki, Dept. of Computer Science, University of Freiburg, Georges-Köhler-Allee 51, D-79110 Freiburg, Germany. E-mail: {zablocki,gausss,lausen}@informatik.uni-freiburg.de Sven Gauß, Dept. of Computer Science, University of Freiburg, Georges-Köhler-Allee 51, D-79110 Freiburg, Germany. E-mail: {zablocki,gausss,lausen}@informatik.uni-freiburg.de Georg Lausen, REWE Information Systems GmbH, Humboldtstraße 140-144, D-51149 Köln, Germany. E-mail: thomas.hornung@rewe-group.com Journal Web Intelligence and Agent Systems Online ISSN 1875-9289 Print ISSN 1570-1263 Journal Volume Volume 12 Journal Issue Volume 12, Number 3 / 2014
Print ISSN:
1570-1263
Electronic ISSN:
1875-9289
Topics:
Computer Science
Permalink