A Hybrid Approach for Prominent Node Detection in Social Media Networks Using Modified Cluster Walktrap with Analytical Hierarchy Process

Abstract

The rapid growth of social media platforms has significantly increased the complexity of online interaction networks, creating new challenges in suspicious community detection and influential node identification. Conventional community detection approaches often face limitations in handling heterogeneous social network structures and identifying structurally significant nodes using multiple decision parameters simultaneously. To address these limitations, this study proposes a hybrid framework integrating modified cluster Walktrap (MCW) with the Analytical Hierarchy Process (AHP) for suspicious community detection and prominent node analysis in social media networks. In the first phase, the proposed MCW approach utilizes probability-driven random-walk analysis and modularity optimization to identify structurally similar communities in bipartite graph structures. In the second phase, AHP is applied to evaluate influential nodes using multiple graph-theoretic parameters, including mean distance, degree centrality, and betweenness centrality. The proposed framework was experimentally evaluated using the publicly available KONECT Facebook social network dataset. Experimental results demonstrate that the proposed MCW + AHP framework achieved improved clustering effectiveness, reduced overlapping community formation, and more structurally consistent community detection compared with conventional Cluster Edge Betweenness and Walktrap approaches. The framework also enabled balanced multi-criteria influential node evaluation, where Node 101 was identified as the most structurally significant node within the analyzed network. Overall, the proposed framework provides a computationally feasible and effective approach for structural community analysis and influential node identification in complex social media environments.

Received: 12 April 2026 | Revised: 23 June 2026 | Accepted: 3 July 2026

Conflicts of Interest

The authors declare that they have no conflicts of interest to this work.

Data Availability Statement

The data that support the findings of this study are openly available in KONECT at https://networkrepository.com/socfb.php.

Author Contribution Statement

Faz Mohammad: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing – original draft, Writing – review & editing, Visualization, Supervision, Project administration. Shweta Vikram: Data curation, Writing – original draft, Writing – review & editing, Visualization, Supervision, Project administration.