DAG: Deep Adaptive and Generative $K$-Free Community Detection on Attributed Graphs

2025-02-20

AI Summary

To address the limitations of existing attributed graph community detection methods—namely, the requirement to pre-specify the number of communities *K* and poor applicability in privacy-sensitive scenarios (e.g., online gaming matchmaking)—this paper proposes K-Free, a novel framework for *K*-free community detection. Methodologically, it introduces: (1) an end-to-end differentiable *K* search mechanism that reformulates discrete hyperparameter grid search over *K* as a continuous optimization problem; (2) a graph neural network integrating masked attribute reconstruction, a community membership readout module, and a group-sparse regularized auto-*K* inference module; and (3) EDGE, a label-free evaluation metric. Extensive experiments on five public benchmarks and real-world Tencent mobile game data demonstrate significant improvements over state-of-the-art methods. Online A/B testing shows a 7.35% gain in team-matching effectiveness.

Technology Category

• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Machine Learning: Graph-based Machine Learning
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

Community detection on attributed graphs with rich semantic and topological information offers great potential for real-world network analysis, especially user matching in online games. Graph Neural Networks (GNNs) have recently enabled Deep Graph Clustering (DGC) methods to learn cluster assignments from semantic and topological information. However, their success depends on the prior knowledge related to the number of communities $K$, which is unrealistic due to the high costs and privacy issues of acquisition.In this paper, we investigate the community detection problem without prior $K$, referred to as $K$-Free Community Detection problem. To address this problem, we propose a novel Deep Adaptive and Generative model~(DAG) for community detection without specifying the prior $K$. DAG consists of three key components, extit{i.e.,} a node representation learning module with masked attribute reconstruction, a community affiliation readout module, and a community number search module with group sparsity. These components enable DAG to convert the process of non-differentiable grid search for the community number, extit{i.e.,} a discrete hyperparameter in existing DGC methods, into a differentiable learning process. In such a way, DAG can simultaneously perform community detection and community number search end-to-end. To alleviate the cost of acquiring community labels in real-world applications, we design a new metric, EDGE, to evaluate community detection methods even when the labels are not feasible. Extensive offline experiments on five public datasets and a real-world online mobile game dataset demonstrate the superiority of our DAG over the existing state-of-the-art (SOTA) methods. DAG has a relative increase of 7.35% in teams in a Tencent online game compared with the best competitor.

Problem

• Detects communities without prior knowledge of K.
• Integrates semantic and topological information for clustering.
• Proposes a differentiable model for community number search.

Innovation

• Deep Adaptive Generative model
• Masked attribute reconstruction
• Differentiable community number search