A Spin-Glass Model for Semi-Supervised Community Detection

• DOI: 10.1609/aaai.v26i1.8320
• Publication Date: 2022-06-01T20:35:18Z
• Authors: Eric Eaton; Rachael Mansbach
• Keywords: Modularity; Benchmark (surveying); PageRank

Current modularity-based community detection methods show decreased performance as relational networks become increasingly noisy. These also yield a large number of diverse structures solutions, which is problematic for applications that impose constraints on the acceptable solutions or in cases where user focused specific communities interest. To address both these problems, we develop semi-supervised spin-glass model enables current to incorporate background knowledge forms individual labels and pairwise constraints. Unlike methods, our approach shows robust presence noise network, ability guide discovery process toward structures. We evaluate algorithm several benchmark new political sentiment network representing cooperative events between nations was mined from news articles over six years.