Graph Neural Networks for Social Network Analysis and Knowledge Graph Completion
Download PDFAbstract
This chapter provides a comprehensive exploration of Graph Neural Networks (GNNs) and their applications in two critical domains: social network analysis and knowledge graph completion. We begin by introducing the foundational concepts of GNNs, including their architectural variants like Graph Convolutional Networks (GCNs), Graph Attention Networks (GATs), and GraphSAGE. The chapter then delves into the practical application of these models for tasks such as community detection and node classification in social networks, using the Cora citation dataset as a case study. Subsequently, we investigate the role of GNNs in knowledge graph completion, focusing on link prediction with the FB15k-237 dataset. A hybrid GNN framework is proposed, integrating multiple architectures to address the distinct challenges of each domain. The chapter presents a detailed methodology, including the experimental setup, training configurations, and evaluation metrics. The results and discussion section provides a thorough analysis of the model’s performance, including comparisons with baseline models and ablation studies. Finally, we conclude with a summary of the key findings and a discussion of future research directions in the field of GNNs.
References
- Zonghan Wu et al. “A comprehensive survey on graph neural networks”. In: IEEE transactions on neural networks and learning systems 32.1 (2020), pp. 4–24.
- TN Kipf. “Semi-supervised classification with graph convolutional networks”. In: arXiv preprint arXiv:1609.02907 (2016).
- Petar Velickovic et al. “Graph attention networks”. In: stat 1050.20 (2017), pp. 10– 48550.
- Will Hamilton, Zhitao Ying, and Jure Leskovec. “Inductive representation learning on large graphs”. In: Advances in neural information processing systems 30 (2017).
- Sadamori Kojaku et al. “Network community detection via neural embeddings”. In: Nature Communications 15.1 (2024), p. 9446.
- Antoine Bordes et al. “Translating embeddings for modeling multi-relational data”. In: Advances in neural information processing systems 26 (2013).
- Bishan Yang et al. “Embedding entities and relations for learning and inference in knowledge bases”. In: arXiv preprint arXiv:1412.6575 (2014).
- Michael Schlichtkrull et al. “Modeling relational data with graph convolutional networks”. In: European semantic web conference. Springer. 2018, pp. 593–607.
- Allam Balaram et al. “Managing 5G IOT Network Operations and Safety Using Deep Learning and Attention Methods”. In: Wireless Personal Communications (2024), pp. 1–16.
- Mohamad Zamini, Hassan Reza, and Minou Rabiei. “A review of knowledge graph completion”. In: Information 13.8 (2022), p. 396.
