Computer Science Location: Rice Hall 204 & Zoom
Add to Calendar 2022-11-17T15:00:00 2022-11-17T15:00:00 America/New_York Ph.D. Proposal Presentation by Nan Wang Building Transparent and Fair Personalization Systems   Abstract: Rice Hall 204 & Zoom

Building Transparent and Fair Personalization Systems

 

Abstract:

Personalization systems (PS) are adopted in nearly every corner of the internet through personalized recommendation, content supply, messaging, and so on. Not only do e-commerce merchants depend on personalization for pushing the relevant items to the right users, but also consumers need personalization to find useful information without being overwhelmed in the info-times. Due to these reasons, enormous efforts have long been devoted in developing more powerful AI and machine learning techniques to improve the performance of PS. In recent years, however, people start to realize that PS empowered by these techniques may lead to undesired effects on users, items, producers, platforms, or even the society at large, which eventually will run counter to the original good purposes of personalization. For example, non-transparency can compromise users’ trust and reliance in the system, unfair treatment on different users or producers may discourage their usage of the system or even raise legal concerns, just to name a few. In this dissertation proposal, I focus on improving the transparency and ensuring the fairness of PS, while maintaining high personalization performance. I believe that transparency and fairness are critical in further expanding the adoption of PS for more efficient and safer applications. For transparency, I propose frameworks for providing users with intuitive textual explanations on the personalization results. The explanations are expected to help users make more informed decisions and build trust in the system. For fairness, I propose algorithms that generate personalized results without discrimination in serving users with different social constructs. The proposed research is expected to help users make faster and better decisions, and improve their trust and satisfaction in PS.

 

Committee

  • Aidong Zhang, Chair, CS/SEAS/UVA
  • Hongning Wang, Advisor, CS/SEAS/UVA
  • Yangfeng Ji, CS/SEAS/UVA
  • Sheng Li, School of Data Science/UVA
  • Yongfeng Zhang, CS/School of Arts and Sciences/Rutgers