Title :
Boosting Response Aware Model-Based Collaborative Filtering
Author :
Haiqin Yang ; Guang Ling ; Yuxin Su ; Lyu, R. ; King, Irwin
Author_Institution :
Dept. of Comput. Sci. & Eng., Chinese Univ. of Hong Kong, Shatin, China
Abstract :
Recommender systems are promising for providing personalized favorite services. Collaborative filtering (CF) technologies, making prediction of users´ preference based on users´ previous behaviors, have become one of the most successful techniques to build modern recommender systems. Several challenging issues occur in previously proposed CF methods: (1) most CF methods ignore users´ response patterns and may yield biased parameter estimation and suboptimal performance; (2) some CF methods adopt heuristic weight settings, which lacks a systematical implementation; and (3) the multinomial mixture models may weaken the computational ability of matrix factorization for generating the data matrix, thus increasing the computational cost of training. To resolve these issues, we incorporate users´ response models into the probabilistic matrix factorization (PMF), a popular matrix factorization CF model, to establish the response aware probabilistic matrix factorization (RAPMF) framework. More specifically, we make the assumption on the user response as a Bernoulli distribution which is parameterized by the rating scores for the observed ratings while as a step function for the unobserved ratings. Moreover, we speed up the algorithm by a mini-batch implementation and a crafting scheduling policy. Finally, we design different experimental protocols and conduct systematical empirical evaluation on both synthetic and real-world datasets to demonstrate the merits of the proposed RAPMF and its mini-batch implementation.
Keywords :
behavioural sciences computing; collaborative filtering; computational complexity; learning (artificial intelligence); matrix decomposition; recommender systems; statistical distributions; Bernoulli distribution; CF methods; RAPMF framework; biased parameter estimation; computational ability; computational cost; data matrix; empirical evaluation; heuristic weight; minibatch learning implementation; multinomial mixture models; observed ratings; rating scores; real-world datasets; recommender systems; response aware model-based collaborative filtering; response aware probabilistic matrix factorization framework; scheduling policy; step function; suboptimal performance; synthetic datasets; unobserved ratings; user behaviors; user preference prediction; user response models; Collaboration; Computational modeling; Data models; Probabilistic logic; Recommender systems; Stochastic processes; Training; Recommender systems; collaborative filtering; matrix factorization; missing data theory;
Journal_Title :
Knowledge and Data Engineering, IEEE Transactions on
DOI :
10.1109/TKDE.2015.2405556