Why You Should Listen to This Song: Reason Generation for Explainable Recommendation

Guoshuai Zhao, Hao Fu, Ruihua Song, Tetsuya Sakai, Xing Xie, Xueming Qian

Abstract: Explainable recommendation, which makes a user aware of why such items are recommended has received a lot of attention as a highly practical research topic. The goal of our research is to make the users feel as if they are receiving recommendations from their friends. To this end, we formulate a new challenging problem called reason generation for explainable recommendation in conversation applications, and propose a solution that generates a natural language explanation of the reason for recommending an item to that particular user. Evaluation with manual assessments indicates that our generated reasons are relevant to songs and personalized to users. They are also fluent and easy to understand. A large-scale online experiments show that our method outperforms manually selected reasons by 8.2% in terms of click-through rate.