Link Predictions in an Online Health Community for Smoking Cessation
Published in KDD Workshop on Mining and Learning with Graphs, 2020
Sulyun Lee, Hankyu Jang, Kang Zhao, Michael S. Amato, and Amanda L. Graham
Abstract: Effective link predictions in online social networks can help to improve user experience and engagement, which are often associated with better health outcomes for users of online health communities (OHCs). However, limited attention has been paid to predicting social network links in OHCs. This paper explores link predictions in an OHC for smoking cessation by considering it as a multi-relational social network that incorporates multiple types of social relationships. We demonstrate that leveraging information from multiple networks built based on different types of relationships is superior to using only information from a single network or the aggregated network. In addition, adding community structures and nodal similarities based on network embedding can help link predictions in different ways. Our work has implications for the design and management of a successful online health community.
Paper link: http://www.mlgworkshop.org/2020/papers/MLG2020_paper_25.pdf
Recommended citation:
Sulyun Lee, Hankyu Jang, Kang Zhao, Michael S. Amato, and Amanda L. Graham. Link Prediction in anOnline Health Community for Smoking Cessation. KDD workshop on Mining and Learning with Graphs, 2020.