Additive co-clustering of gaussians and poissons for joint modeling of ratings and reviews
NIPS workshop on Nonparametric Methods for Large Scale Representation Learning(2015)
摘要
Understanding a user’s motivations provides valuable information beyond the ability to recommend items. Quite often this can be accomplished by perusing both ratings and review texts, since it is the latter where the reasoning for specific preferences is explicitly expressed.Unfortunately matrix factorization approaches to recommendation result in large, complex models that are difficult to interpret and give recommendations that are hard to clearly explain to users. In contrast, in this paper, we attack this problem through succinct additive co-clustering. We devise a novel Bayesian technique for summing co-clusterings of joint Gaussian and Poisson distributions. With this novel technique we propose a new Bayesian model for joint collaborative filtering of ratings and text reviews through a sum of simple co-clusterings. Our model is non-parametric in the size of the co-clusterings and we offer a novel efficient sampling algorithm for learning the sum of Poisson distributions and the sum of Gaussian distributions. The simple structure of our model yields easily interpretable recommendations. Even with a simple, succinct structure, our model outperforms competitors in terms of predicting both ratings and reviews.
更多查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络