S. Srividhya,
Dr. Lavanya S.R,
- Associate Professor,, KPR College of Arts Science and Research Coimbatore, Uthupalayam,, Tamil Nadu,, India
- Associate Professor,, KPR College of Arts Science and Research Coimbatore, Uthupalayam,, Tamil Nadu,, India
Abstract
Community detection in complex networks is a fundamental problem with applications across diverse domains, ranging from social networks to biological systems and beyond. Traditional methods based on graph theory have been widely used for identifying communities within networks. However, the intricate and evolving nature of modern networks demands more sophisticated approaches. This research paper proposes a hybrid approach that combines the strengths of deep learning techniques with traditional community detection algorithms for improved accuracy and scalability. Leveraging the capabilities of deep learning models, specifically Graph Neural Networks (GNNs), the proposed approach aims to learn intricate patterns and representations from network data, encoding rich structural information into node embeddings. The first phase involves employing Graph Neural Networks, such as Graph Convolutional Networks (GCNs), to learn node representations that encapsulate complex relationships and local structures within the network. These learned embeddings capture both topological information and node attributes, enhancing the understanding of the network topology. Subsequently, the learned node embeddings are integrated into traditional community detection algorithms, such as spectral clustering to form a hybrid framework. This integration allows for the utilization of deep learning-based representations within established community detection methodologies, aiming to enhance the accuracy and robustness of community detection. The proposed hybrid approach is evaluated on various benchmark datasets and real-world networks to assess its effectiveness in detecting communities. Performance metrics such as modularity, conductance, and coverage are utilized to evaluate the quality of detected communities compared to standalone traditional or deep learning-based methods. The results demonstrate the efficacy of the hybrid approach, showcasing improved modularity, conductance, coverage and runtime compared to individual methods. Furthermore, the paper discusses the significance of leveraging both deep learning techniques and traditional algorithms in community detection tasks.
Keywords: Community Detection, Deep Learning, Graph Neural Network, Hybrid Approach, Graph Convolution Network
[This article belongs to Recent Trends in Electronics Communication Systems (rtecs)]
S. Srividhya, Dr. Lavanya S.R. Hybrid Approach for Community Detection Using Deep Learning Techniques. Recent Trends in Electronics Communication Systems. 2024; 11(03):1-10.
S. Srividhya, Dr. Lavanya S.R. Hybrid Approach for Community Detection Using Deep Learning Techniques. Recent Trends in Electronics Communication Systems. 2024; 11(03):1-10. Available from: https://journals.stmjournals.com/rtecs/article=2024/view=176497
References
- Newman, M. E. J. (2010). Networks: An Introduction. Oxford University Press. England https://doi.org/10.1093/acprof:oso/9780199206650.001.0001,
- Fortunato, S. (2010). Community detection in graphs. Physics Reports, 486(3-5), 75-174.
- Newman, M. E. J., & Girvan, M. (2004). Finding and evaluating community structure in networks. Physical Review E, 69(2), 026113.
- Von Luxburg, U. (2007). A tutorial on spectral clustering. Statistics and Computing, 17(4), 395-416.
- Rosvall, M., & Bergstrom, C. T. (2008). Maps of random walks on complex networks reveal community structure. Proceedings of the National Academy of Sciences, 105(4), 1118-1123.
- Wu, Z., Pan, S., Long, G., Jiang, J., Zhang, C., & Yu, P. S. (2021). A comprehensive survey on graph neural networks. IEEE Transactions on Neural Networks and Learning Systems, 32(1), 4-24.
- Kipf, T. N., & Welling, M. (2016). Semi-supervised classification with graph convolutional networks. International Conference on Learning Representations (ICLR).
- Hamilton, W. L., Ying, R., & Leskovec, J. (2017). Inductive representation learning on large graphs. Advances in Neural Information Processing Systems (NeurIPS).
- Veličković, P., Cucurull, G., Casanova, A., Romero, A., Lio, P., & Bengio, Y. (2018). Graph attention networks. International Conference on Learning Representations (ICLR).
- Perozzi, B., Al-Rfou, R., & Skiena, S. (2014). DeepWalk: Online learning of social representations. Proceedings of the 20th ACM SIGKDD international conference on Knowledge discovery and data mining, New York New York USA August 24 – 27, 2014; 701-710.
- Wang, D., Cui, P., & Zhu, W. Structural deep clustering networks for unsupervised learning. Proceedings of the 27th ACM International Conference on Information and Knowledge Management, Taipei Taiwan April 20 – 24, 2020; ACM digital library. 1949-1952.
- Guo R, Zou J, Bai Q, Wang W, Chang X. Community Detection Fusing Graph Attention Network. Mathematics. 2022; 10(21):4155. https://doi.org/10.3390/math10214155
- Aditya Grover and Jure Leskovec. 2016. Node2vec: Scalable Feature Learning for Networks. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD ’16). Association for Computing Machinery, New York, NY, USA, 855–864. https://doi.org/10.1145/2939672.2939754
- Newman, M. E. J., & Girvan, M. (2004). Finding and evaluating community structure in networks. Physical Review E, 2004; 69(2), 026113. https://doi.org/10.1103/PhysRevE.69.026113
Recent Trends in Electronics Communication Systems
| Volume | 11 |
| Issue | 03 |
| Received | 23/07/2024 |
| Accepted | 10/08/2024 |
| Published | 01/10/2024 |