A Novel Movie Recommendation System Based on Deep Reinforcement Learning with Prioritized Experience Replay
Zhang Yuyan, Su Xiayao, Yong Liu · 2019
A recommendation system plays an important role in information overload case by recommending personalized services to improve user experience. In this paper, a novel movie recommendation system based on deep reinforcement learning (DRL) framework is proposed. In proposed system model, the state information is preprocessed to overcome the problems of data sparsity and cold start. Specially, user's interest change is captured using cross entropy and used to prioritize experience replay in a replay memory. The experiments verify that the proposed model can speed up the network update and improve recommendation accuracy.