Contextual Recommendation based on Text Mining

摘要

The potential benefit of integrating con-textual information for recommendation has received much research attention re-cently, especially with the ever-increasing interest in mobile-based recommendation services. However, context based recom-mendation research is limited due to the lack of standard evaluation data with con-textual information and reliable technol-ogy for extracting such information. As a result, there are no widely accepted con-clusions on how, when and whether con-text helps. Additionally, a system of-ten suffers from the so called cold start problem due to the lack of data for train-ing the initial context based recommenda-tion model. This paper proposes a novel solution to address these problems with automated information extraction tech-niques. We also compare several ap-proaches for utilizing context based on a new data set collected using the pro-posed solution. The experimental results demonstrate that 1) IE-based techniques can help create a large scale context data with decent quality from online reviews, at least for restaurant recommendations; 2) context helps recommender systems rank items, however, does not help pre-dict user ratings; 3) simply using context to filter items hurts recommendation per-formance, while a new probabilistic latent relational model we proposed helps.