Topic Cluster Identification in Smart Grid Cybersecurity Research Using Spectral Clustering

The rapid integration of communication technologies into modern power systems has transformed conventional grids into smart grids, offering enhanced efficiency, reliability, and sustainability. However, this increased connectivity also introduces significant cybersecurity challenges, making it crucial to understand research trends and thematic clusters within this field. Identifying such clusters provides insights into emerging areas, research gaps, and opportunities for developing more resilient smart grid infrastructures. Despite a growing body of literature, smart grid cybersecurity research remains highly fragmented, with topics ranging from intrusion detection to cryptographic protocols and privacy-preserving mechanisms. Traditional literature reviews often struggle to capture the structural relationships across these domains. Consequently, there is a need for systematic approaches that can uncover hidden patterns and effectively classify research topics. This paper proposes a Spectral Clustering-based Topic Cluster Identification Framework (SCTCIF) to analyze smart grid cybersecurity literature. The framework utilizes spectral clustering to analyze bibliometric networks derived from co-occurrence data of keywords and abstracts, thereby enabling the identification of cohesive research communities. SC-TCIF leverages eigenvalue decomposition of the graph Laplacian to identify clusters that traditional clustering methods might overlook, ensuring more precise topic delineation. Experimental results on a dataset of recent publications demonstrate that SC-TCIF effectively identifies distinct topic clusters such as intrusion detection, blockchain-based security, privacy protection, and secure communication protocols. Compared to baseline clustering techniques, SC-TCIF achieves higher modularity and coherence scores. In conclusion, the proposed framework offers a robust analytical tool for mapping smart grid cybersecurity research, enabling scholars and practitioners to navigate the field and identify emerging research directions.

Ҳали таржима қилинмаган