Feature-Based Link Predictions Under Graph Root Distribution

• Activity Number: 171 - SPAAC Poster Competition
• Type: Topic Contributed
• Date/Time: Tuesday, August 4, 2020 : 10:00 AM to 2:00 PM
• Sponsor: Section on Statistical Learning and Data Science
• Author(s): Jiashen Lu* and Kehui Chen
• Companies: University of Pittsburgh and University of Pittsburgh
• Keywords: Link Prediction; Non-parametric; Network Data

Abstract

User-user interaction data is common nowadays, for example, we can observe friendship relations on Facebook or existed links between different websites. So the natural question we can ask is, for a new person, could we make personalized recommendations for possible links based on his/her unique feature? In this project, we build a new framework using Graph Root Distribution trying to answer such questions. We investigate the performance of our method based on several real data sets.