Academic Report of School of Mathematical Sciences [2025] No. 155

(Series Report for High-Level University Construction No. 1256)

Title：Community Detection in Heterogeneous Symbol Networks

Speaker：Associate Professor Jingnan Zhang (University of Science and Technology of China)

Time：16:00-17:00, Dec.26, 2025

Location：Room 307, Alumni Plaza

Abstract： In recent years, network data has garnered increasing attention across scientific disciplines, driving the development of diverse network models. Existing network analysis methods primarily target unsigned networks, while signed networks with both positive and negative edges frequently appear in practice yet remain relatively understudied. This paper formally defines strong equilibrium and weak equilibrium in signed networks and proposes a signed block β-model capable of modeling both strong and weak equilibrium signed networks. Leveraging bipartite graph properties, we establish the identifiability of the proposed model and develop an efficient alternating update algorithm to optimize the resulting log-likelihood function. Crucially, we prove asymptotic consistency for both probability estimation and community detection. Extensive numerical experiments and applications to real-world international relations networks further validate the method's superiority.

Speaker Profile：Dr. Jingnan Zhang is currently an Associate Professor in the Department of Statistics and Finance at the School of Management, University of Science and Technology of China. He received his Ph.D. from City University of Hong Kong. His primary research focuses on network data analysis and related issues. He has published multiple papers in statistical journals such as JASA, Biometrics, JCGS, and NeurIPS, as well as artificial intelligence conferences. He currently holds one National Natural Science Foundation of China grant.

Faculty and students are welcome to attend!

Invited by: Jiangzhou Wang  

School of Mathematical Sciences

December 12, 2025