Recommender Systems on Platforms (Joint work with Xuan Wang and Dennis J. Zhang)
报告人： Renyu (Philip) Zhang, New York University Shanghai
地点： K02 教室,光华2号楼
We study how a recommender system could shape the demand to maximize the number of matches between consumers and producers on a two-sided platform. To improve the long-term performance of a platform, the optimal recommender system design trades off expanding market demand by recommending high-quality producers and reducing congestion by recommending low-quality ones. The traditional recommender system has been focusing on identifying and recommending the producers of highest qualities. Albeit optimal when producers have unlimited capacities, such a recommender system may result in substantial optimality losses in the presence of producer capacity constraints. Such losses are most significant when the demand and supply of the platform is balanced. If, in addition, the quality of a producer improves upon working for the platform, the platform should recommend low-quality producers even more often to improve their qualities. A dynamic re-balancing recommendation policy which maximizes the number of matches myopically converges to an optimal policy in the long-run. We also extend our results to a setting with multiple quality levels and a setting with horizontally heterogeneous consumers and producers. Our results shed light on the optimal recommender system design for a broad range of two-sided platforms.
About the Speaker:
Renyu (Philip) Zhang has been an Assistant Professor of Operations Management at New York University Shanghai since August 2016. His research addresses fundamental operations issues under the emerging trends in technology, marketplaces, and society. He is particularly enthusiastic about developing analytics techniques to study the operations problems in the context of online platforms and marketplaces, sharing economy, social networks, and sustainability. His research works have appeared in Operations Research and Manufacturing & Service Operations Management. Please visit his personal website for more about Philip: https://www.nyu.edu/projects/rzhang/. Before joining NYU Shanghai, Philip obtained his doctoral degree in Operations Management at Olin Business School, Washington University in St. Louis in May 2016 under the supervision of Professor Nan Yang and Professor Fuqiang Zhang.