Title: Product Recommendations Based on Comments: Discovering What You Like Through Strangers

Speaker: Chenliang Li

Time: Friday, January 18, 2019 10:30-11:30

Venue: C403, Dingxin Building, Central Campus of Jilin University

Organizer: School of Artificial Intelligence

Abstract

The purpose of the recommendation system is to provide users with personalized recommendations to increase user dependence and satisfaction with the platform. However, the sparseness of user-commodity interaction behavior data often leads to unsatisfactory recommendations. In recent years, the recommendation model based on user comments has become a research hotspot in the field of recommendation. However, the user's comment text itself will also face the problem of insufficient information, such as insufficient comments and short comments. This report will introduce a work to improve the accuracy of "product score" prediction by mining the user's comment data. For a given user-item pair, this work can extract features from the auxiliary comments written by other users with similar interests of the current user, thus to better improve the quality of the recommendation.

Short biography of Dr. Chenliang Li

Chenliang Li, Ph.D., is currently an associate professor of the School of National Cyber Security of Wuhan University, and a "Luojia honor young scholar" of Wuhan University. He is a member of the Youth Work Committee of the Chinese Information Processing Society of China, a member of the Social Media Special Committee, and a member of the Information Retrieval Committee. He is a reviewer of top computer science journals such as IEEE TKDE, ACM TOIS, and JASIST, and a Program Committee Member for conferences such as SIGIR, ACL, CIKM, WWW, AAAI, and IJCAI. He is a member of the editorial board of JASIST. His research interests include information retrieval, natural language processing, machine learning, and social media analysis. He has published nearly 30 papers in top conferences and journals such as TKDE, TOIS, SIGIR, ACL, AAAI, CIKM, and JASIST. He was awarded the SIGIR 2016 Best Student Paper Award Honorable Mention, and SIGIR 2017 Outstanding Reviewer Award.