Title: Review-based Product Recommendations: Know What You Like Through Strangers
Speaker: Chenliang Li
Time: 10: 30-11: 30, Friday, January 18, 2019
Venue: Room C403, Dingxin Building, Jilin University
Organizer: School of Artificial Intelligence
Abstract:
As we all know, the purpose of a recommendation system is to provide users with personalized recommendations to improve the user’s reliance on the platform and satisfaction. However, the sparsity of the interaction data between users and products often leads to unsatisfactory recommendation results. In recent years, the recommendation model based on users’ comments has become a research topic in the field of recommendation. However, we also face the problem of the insufficient information towards the users’ comments text, including the phenomenon of insufficient number of comments and too brief comments. This report will introduce a method to improve the accuracy of product rating prediction by mining peer review data. This work can be done to improve the quality of recommendations by extracting features for current user and product pairs from helpful comments written by like-minded users who have similar interests to the current users.
Biography:
Chenliang Li is an Associate Professor of School of Cyber Science and Engineering-WHU, Master’s Candidates Supervisor, and a young scholar of Luojia. He is a member of the Youth Working Committee, the Social Media Committee, and the Information Retrieval Committee of the Chinese Information Processing Society of China. He serves as a reviewer of important international academic journals such as IEEE TKDE, ACM TOIS, and JASIST. In addition, he is a Member of the TPC of SIGIR, ACL, CIKM, WWW, AAAI, and IJCAI. He serves as the editorial board of JASIST. His research interests include Information Retrieval, Natural Language Processing, Machine Learning, and Social Media Analysis. Nearly 30 papers have been published in authoritative conferences and journals such as TKDE, TOIS, SIGIR, ACL, AAAI, CIKM, and JASIST. He won the Best Student Paper Award Honorable Mention of SIGIR 2016 and the Outstanding Reviewer Award of SIGIR 2017.
