New Recipe for Semi-supervised Community Detection: Clique Annealing under Crystallization Kinetics

2025-04-22

AI Summary

In label-scarce settings, existing semi-supervised community detection methods suffer from suboptimal initial core selection and poor scalability due to reliance on complex, computationally expensive models. To address this, this paper pioneers an analogy between community evolution and crystallization dynamics, proposing a novel “spontaneous nucleation and annealing-based growth” paradigm. Our key contributions are: (1) the first application of crystallization kinetics principles to guide semi-supervised community discovery; (2) a learning-free transitive annealer that enables adaptive, low-overhead, and scalable optimization of community cores; and (3) integration of clique-based annealing with transitive clique merging. Evaluated across 43 network configurations and multiple real-world benchmarks, our method consistently outperforms state-of-the-art approaches in both accuracy and efficiency, achieving superior scalability without sacrificing precision.

Technology Category

• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Machine Learning: Semi-Supervised Learning
• Search and Optimization: Evolutionary Computation

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Web Mining and Content Analysis: Models for Web evolution
• Social Networks and Social Media: Social mining and social search on the Web

Abstract

Semi-supervised community detection methods are widely used for identifying specific communities due to the label scarcity. Existing semi-supervised community detection methods typically involve two learning stages learning in both initial identification and subsequent adjustment, which often starts from an unreasonable community core candidate. Moreover, these methods encounter scalability issues because they depend on reinforcement learning and generative adversarial networks, leading to higher computational costs and restricting the selection of candidates. To address these limitations, we draw a parallel between crystallization kinetics and community detection to integrate the spontaneity of the annealing process into community detection. Specifically, we liken community detection to identifying a crystal subgrain (core) that expands into a complete grain (community) through a process similar to annealing. Based on this finding, we propose CLique ANNealing (CLANN), which applies kinetics concepts to community detection by integrating these principles into the optimization process to strengthen the consistency of the community core. Subsequently, a learning-free Transitive Annealer was employed to refine the first-stage candidates by merging neighboring cliques and repositioning the community core, enabling a spontaneous growth process that enhances scalability. Extensive experiments on extbf{43} different network settings demonstrate that CLANN outperforms state-of-the-art methods across multiple real-world datasets, showcasing its exceptional efficacy and efficiency in community detection.

Problem

• Improving semi-supervised community detection accuracy
• Reducing computational costs in community detection
• Enhancing scalability of community detection methods

Innovation

• Clique Annealing mimics crystallization kinetics
• Learning-free Transitive Annealer refines candidates
• Integrates annealing principles for core consistency