A Synthetic Approach for Recommendation: Combining Ratings, Social\n Relations, and Reviews

Guangneng Hu, Xinyu Dai, Yunya Song, Shujian Huang, Jiajun Chen

arXiv (Cornell University) · 2016 · 37 citations · 29 references

DOIFull text

Open access

Concepts

Abstract

Recommender systems (RSs) provide an effective way of alleviating the\ninformation overload problem by selecting personalized choices. Online social\nnetworks and user-generated content provide diverse sources for recommendation\nbeyond ratings, which present opportunities as well as challenges for\ntraditional RSs. Although social matrix factorization (Social MF) can integrate\nratings with social relations and topic matrix factorization can integrate\nratings with item reviews, both of them ignore some useful information. In this\npaper, we investigate the effective data fusion by combining the two\napproaches, in two steps. First, we extend Social MF to exploit the graph\nstructure of neighbors. Second, we propose a novel framework MR3 to jointly\nmodel these three types of information effectively for rating prediction by\naligning latent factors and hidden topics. We achieve more accurate rating\nprediction on two real-life datasets. Furthermore, we measure the contribution\nof each data source to the proposed framework.\n

References

29