【Job talk】Recommender Systems and the Value of User Data

  • 研討會日期 : 2021-12-24
  • 時間 : 10:00
  • 主講人 : Dr. Gunhaeng Lee
  • 主持人 : Professor Tzu-Ting Yang
  • 地點 : Online or Conference Room B110
  • 演講者簡介 : Dr. Lee received his Ph.D. in Economics from the Pennsylvania State University in 2018. He is currently the Postdoctoral research fellow at National University of Singapore. His research interests are Applied microeconomics theory, Microeconomic theory, and Industrial organization. He is applying for a position of the Institute of Economics, Academia Sinica now.
  • 演講摘要 : Many online platforms and other technology-enabled firms including the likes of Amazon, Goodreads, LinkedIn, Netflix, StitchFix, Tinder and YouTube make use of some type of recommender system to help their users discover relevant products, people, or content, thereby increasing the value of the services they offer. In this paper we model how a recommender system works, and use the model to investigate, both theoretically and empirically, what is the value provided by user data. Value is created in our setting by customization, selection and screening. We determine how the value of data varies with the number of target items available for the user, the number of users that data is gathered from, the number of features that are taken into account in the customized prediction, and the degree of misalignment between the platform’s and user’s interests.