B1255
Title: Supervised centrality via sparse spatial autoregression
Authors: Chenlei Leng - University of Warwick (United Kingdom) [presenting]
Abstract: The social opinions, behaviors and sentiments of the players in a social network are closely associated with their network positions. Identifying the influential players in a network is of importance as it helps to understand how ties are formed, how information is propagated, and in turn, can guide the dissemination of new information by focusing on important players. Motivated by a Weibo social network on 2021 Henan Floods, where response variables on each node are available, we propose a novel notion of supervised centrality to account for the fact that the centrality of a node is task-specific. To estimate the supervised centrality and identify important players, we develop a novel sparse spatial autoregression model by introducing individual heterogeneity to each user. To overcome the computational difficulties with fitting the model for large social networks, we further develop a forward-addition algorithm and show that it can consistently identify a superset of the influential nodes. We apply our model to analyze three responses in the Henan Floods data: the number of comments, reposts and likes, and obtain interesting results. A simulation study further corroborates the developed theory.
