Link Predictions for Social Networks in Online Health Communities


The research project was presented at INFORMS Annual Meeting in a session titled, “Learning and Optimization in Social Media.”

Authors: Sulyun Lee, Hankyu Jang, Kang Zhao, Michael S. Amato and Amanda L. Graham

Abstract: Online Health Communities (OHCs) are popular sources for patients and their families to get informational and emotional support related to the diseases or symptoms. As users interact with each other, they also form a social network. We try to predict the links that are likely to be formed in such a network to improve patients’ experience in an OHC. Our model considers the different types of relationships among users, users’ levels of online activities, as well as the topics of their online discussions. We show that our model outperforms benchmark methods in predicting future social network links.

Talk info link:!/6818/presentation/5441

Presentation slides: slides