Conversational Recommender System Based on GRU-attention Neural Network

In recent years, the conversational recommender system (CRS) based on natural language processing technology has gained widespread attention, aiming to learn and model user preferences through interactive dialogue. Although existing research has improved the accuracy of the dialogue recommendation system to a certain extent, there are still some shortcomings that make it easy to generate more general and popular responses. This paper proposes a deep learning framework with GRU and attention mechanism on the basis of knowledge graph, namely KGGA. Through the serialized modeling of conversation content and effectively capture the user’s interest to make correct recommendations. Experiments conducted on widely baselines show that the proposed framework is superior to the state-of-the-art methods.