CommTrustSVD: A Recommendation Algorithm Based on Comment Mining and Implicit Trust
The information on e-commerce platforms is vast and disordered, so how to find useful knowledge and make accurate recommendation has become an urgent problem.User comments represent the group wisdom of the users, which can reflect the users feelings towards the product.And the implicit trust obtained through users past ratings or interaction can reflect users preferences 1.Considering comments and implicit trust comprehensively, products with good reputation and conforming to users interests will be recommended.Therefore, a recommendation algorithm that considers both the comments and implicit trust of users is proposed.On one hand, the comments information is mined to obtain the comment rating of each product.On the other hand, the implicit trust is used for personalized recommendation.Experiments on Amazon audio equipment dataset demonstrate that the proposed algorithm can improve the accuracy of prediction results and make recommendations for users effectively.
e-commerce comment mining natural language processing implicit trust personalized recommendation
Jialei Wang Yanmei Zhang Xiaoyi Tang
Information School,Central University of Finance and Economics, Beijing 100081, China
国际会议
郑州
英文
156-163
2018-09-21（万方平台首次上网日期，不代表论文的发表时间）