Balancing prediction and recommendation accuracy: Hierarchical latent factors for preference data

Recent works in Recommender Systems (RS) have in- vestigated the relationships between the prediction ac- curacy, i.e. the ability of a RS to minimize a cost func- Tion (for instance the RMSE measure) in estimating users' preferences, and the accuracy of the recommenda- Tion list provided to users. State-of-the-art recommen- dation algorithms, which focus on the minimization of RMSE, have shown to achieve weak results from the rec- ommendation accuracy perspective, and vice versa. In this work we present a novel Bayesian probabilistic hi- erarchical approach for users' preference data, which is designed to overcome the limitation of current method- ologies and thus to meet both prediction and recommen- dation accuracy. According to the generative semantics of this technique, each user is modeled as a random mix- Ture over latent factors, which identify users community interests. Each individual user community is then mod- eled as a mixture of topics, which capture the prefer- ences of the members on a set of items. We provide two different formalization of the basic hierarchical model: BH-Forced focuses on rating prediction, while BH-Free models both the popularity of items and the distribu- Tion over item ratings. The combined modeling of item popularity and rating provides a powerful framework for the generation of highly accurate recommendations. An extensive evaluation over two popular benchmark datasets reveals the effectiveness and the quality of the proposed algorithms, showing that BH-Free realizes the most satisfactory compromise between prediction and recommendation accuracy with respect to several state- of-the-art competitors. Copyright © 2012 by the Society for Industrial and Applied Mathematics.