EDMF: Efficient Deep Matrix Factorization With Review Feature Learning for Industrial Recommender System
Hai Liu, Chao Zheng, Duantengchuan Li +6 more
Recommendation accuracy is a fundamental problem in the quality of the recommendation system. In this article, we propose an efficient deep matrix factorization (EDMF) with review feature learning for the industrial recommender system. Two characteristics in user’s review are revealed. First, interactivity between the user and the item, which can also be considered as the former’s scoring behavior on the latter, is exploited in a review. Second, the review is only a partial description of the user’s preferences for the item, which is revealed as the sparsity property. Specifically, in the first characteristic, EDMF extracts the interactive features of onefold review by convolutional neural networks with word-attention mechanism. Subsequently, <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math…