PhD, Carnegie Mellon University
Email
Abstract
Identify, Recommend, and Configure Social Roles in Online Communities
Millions of volunteers participate in the online communities, exchange their ideas and collaborate to produce complex artifacts. A better understanding of members’ roles can facilitate a variety of downstream applications such as task recommendation. However, identifying social roles in online groups has not been systematically investigated since there lacks a proper definition of roles and systematic methodologies for identifying roles, its evolution, and manifestation under different contexts. To address these challenges, I propose to combine NLP techniques and social science theories to model roles specifically via working in three areas: (1) designing fine-grained taxonomy and machine learning models to capture users’ actions, intentions and roles when participating in groups, (2) modeling the mobility and stability of users’ roles over time, and (3) building interventions such as recommender system to match users to roles and tasks to improve online production communities.
Bio
Diyi Yang is a Ph.D student in the Language Technologies Institute of Carnegie Mellon University, advised by Prof. Robert Kraut and Prof. Eduard Hovy. Her research interests lie in computational social science and natural language processing. Specifically, Diyi’s work focuses utilizing conversational text to study new social problems and design new methods to better understand human behavior, and have been published in leading NLP/HCI conferences. She has been awarded Facebook Fellowship (2017-2019) and Presidential Fellowship, CMU (2016-2017). Prior to CMU, Diyi received her bachelor degree from Shanghai Jiao Tong University.