A RECOMMENDATION MODEL COMBINING GATING ATTENTION MECHANISM AND BILINEAR FEATURE INTERACTION
In order to accurately express the users'true preferences in a single type of recommendation such as movies and books,and fully capture the effective features in the recommendation data,a recommendation model that integrates the gating attention mechanism and bilinear feature interaction is proposed.This model used the attention unit integrated into the gated mechanism to model the user's local explicit preferences and used bilinear feature interaction layer to mine the long-term general preferences of users to improve the learning ability of the deep recommendation model.Experiments were conducted on two public data sets,Amazon(Books)and MovieLens-1M.The experimental results show that the proposed model has a certain degree of improvement in recommendation effect compared with other recommendation models.
国家科技期刊平台
NSTL
万方数据
维普
