Core Triangle Motif-Driven Local Community Detection Algorithm

Community detection, an important research topic in complex network analysis, aims to discover the hidden structure and information within networks. The rapid growth in network size and the increasingly complex structures of networks have placed higher demands on community detection algorithms. To address this issue, this article proposes a core triangle motif-driven local community detection method named CoreTriM. CoreTriM initially selects target nodes based on their descending local importance and then obtains their corresponding core triangle motifs. Next, it utilizes three strategies to identify stable initial communities: label homogeneous transition, label diffusion within core triangle motifs, and label diffusion via local expansion of core triangle motifs. Finally, a novel merging method based on triangle motifs is employed to integrate weak communities, yielding final communities, effectively addressing the issue of limited resolution. Experiments on real-world and synthetic networks demonstrate the superiority of CoreTriM algorithm in terms of accuracy, robustness, execution time, and convergence speed, while its parameter is easy to set.