Gediminas Adomavicius, and YoungOk Kwon

[Full Paper in pdf]

While traditional single-rating recommender systems have been successful in a number of personalization applications, the research area of multi-criteria recommender systems has been largely untouched. In order to take full advantage of multi-criteria ratings in various applications, new recommendation techniques are required. In this paper we propose two new approaches – the similarity-based approach and the aggregation function-based approach – to incorporating and leveraging multi-criteria rating information in recommender systems. We also discuss multiple variations of each proposed approach, and perform empirical analysis of these approaches using a real-world dataset. Our experimental results show that multi-criteria ratings can be successfully leveraged to improve recommendation accuracy, as compared to traditional single-rating recommendation techniques.

KEYWORDS: Personalization, recommender systems, collaborative filtering, multi-criteria ratings, rating estimation.