Applying probabilistic latent semantic analysis to multi-criteria recommender system

Article type: Research Article

Authors: Zhang, Yin | Zhuang, Yueting | Wu, Jiangqin | Zhang, Liang

Affiliations: College of Computer Science, Zhejiang University, Hangzhou 310027, China. E-mails: {zhangyin98, yzhuang, wujq}@cs.zju.edu.cn, cddb.zhang@gmail.com

Abstract: Nowadays some recommender system researchers have already been engaging multi-criteria that model possible attributes of the item to generate the improved recommendations. However, the statistical machine learning methods successful in the single-rating recommender system have not been investigated in the context of multi-criteria ratings. In this paper, we propose two types of multi-criteria probabilistic latent semantic analysis algorithms extended from the single-rating version. First, the mixture of multi-variate Gaussian distribution is assumed to be the underlying distribution of multi-criteria ratings of each user. Second, we further assume the mixture of the linear Gaussian regression model as the underlying distribution of multi-criteria ratings of each user, inspired by the Bayesian network and linear regression. The experiment results on the Yahoo!Movies ratings data set show that the full multi-variate Gaussian model and the linear Gaussian regression model achieve a stable performance gain over other tested methods.

Keywords: Collaborative filtering, multi-criteria, latent semantic analysis, linear Gaussian regression

DOI: 10.3233/AIC-2009-0446

Journal: AI Communications, vol. 22, no. 2, pp. 97-107, 2009