Semantically-enhanced pre-filtering for context-aware recommender systems

Victor Codina, Francesco Ricci, Luigi Ceccaroni
2013 Proceedings of the 3rd Workshop on Context-awareness in Retrieval and Recommendation - CaRR '13  
6HYHUDO UHVHDUFK ZRUNV KDYH GHPRQVWUDWHG WKDW LI XVHUV ¶ UDWLQJV are truly context-dependent, then Context-Aware Recommender Systems can outperform traditional recommenders. In this paper we present a novel contextual pre-filtering approach that exploits the implicit semantic similarity of contextual situations. For GHWHUPLQLQJ VXFK D VLPLODULW\ ZH UHO\ RQO\ RQ WKH DYDLODEOH XVHUV ¶ ratings and we deem as similar two syntactically different contextual situations that are actually influencing in
more » ... a similar way WKH XVHU ¶V UDWLQJ EHKDYLRU :H YDOLGDWH WKH SURSRVHG DSSURDFK using two contextually tagged ratings data sets showing that it outperforms a traditional pre-filtering approach and a state-of-theart context-aware Matrix Factorization model.
doi:10.1145/2442670.2442674 fatcat:ofxjnqoocjecpbpjg47r6yn6yu