Advancing Community Detection with Graph Convolutional Neural Networks: Bridging Topological and Attributive Cohesion

de Silva, Anjali; Chen, Gang; Ma, Hui; Nekooei, Seyed Mohammad; Zuo, Xingquan

doi:10.24963/ijcai.2025/821

Advancing Community Detection with Graph Convolutional Neural Networks: Bridging Topological and Attributive Cohesion

Anjali de Silva, Gang Chen, Hui Ma, Seyed Mohammad Nekooei, Xingquan Zuo

Download BibTeX Cite

Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence. Main Track. Pages 7383-7391.

https://doi.org/10.24963/ijcai.2025/821

Keywords

Multidisciplinary Topics and Applications: MTA: Web and social networks
Machine Learning: ML: Supervised Learning
Machine Learning: ML: Deep learning architectures
Machine Learning: ML: Clustering

Abstract

Community detection, a vital technology for real-world applications, uncovers cohesive node groups (communities) by leveraging both topological and attribute similarities in social networks. However, existing Graph Convolutional Networks (GCNs) trained to maximize modularity often converge to suboptimal solutions. Additionally, directly using human-labeled communities for training can undermine topological cohesiveness by grouping disconnected nodes based solely on node attributes. We address these issues by proposing a novel Topological and Attributive Similarity-based Community detection (TAS-Com) method. TAS-Com introduces a novel loss function that exploits the highly effective and scalable Leiden algorithm to detect community structures with global optimal modularity. Leiden is further utilized to refine human-labeled communities to ensure connectivity within each community, enabling TAS-Com to detect community structures with desirable trade-offs between modularity and compliance with human labels. Experimental results on multiple benchmark networks confirm that TAS-Com can significantly outperform several state-of-the-art algorithms.

Anjali de Silva, Gang Chen, Hui Ma, Seyed Mohammad Nekooei, and Xingquan Zuo. 2025. Advancing Community Detection with Graph Convolutional Neural Networks: Bridging Topological and Attributive Cohesion. In Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence, IJCAI-25. IJCAI, 7383–7391. https://doi.org/10.24963/ijcai.2025/821

@inproceedings{ijcai2025p821,
  title     = {Advancing Community Detection with Graph Convolutional Neural Networks: Bridging Topological and Attributive Cohesion},
  author    = {de Silva, Anjali and Chen, Gang and Ma, Hui and Nekooei, Seyed Mohammad and Zuo, Xingquan},
  booktitle = {Proceedings of the Thirty-Fourth International Joint Conference on
               Artificial Intelligence, {IJCAI-25}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {James Kwok},
  pages     = {7383--7391},
  year      = {2025},
  month     = {8},
  note      = {Main Track},
  doi       = {10.24963/ijcai.2025/821},
  url       = {https://doi.org/10.24963/ijcai.2025/821},
}

TY  - CPAPER
TI  - Advancing Community Detection with Graph Convolutional Neural Networks: Bridging Topological and Attributive Cohesion
AU  - de Silva, Anjali
AU  - Chen, Gang
AU  - Ma, Hui
AU  - Nekooei, Seyed Mohammad
AU  - Zuo, Xingquan
T2  - Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence, IJCAI-25
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2025
SP  - 7383
EP  - 7391
DO  - 10.24963/ijcai.2025/821
UR  - https://doi.org/10.24963/ijcai.2025/821
ER  -