Enhanced Identification of Influential Nodes in Social Networks Using an Optimized Fuzzy Clustering and SALT-Based Algorithm

Abstract

Identifying influential social network nodes is crucial in today's interconnected world for supporting strategic decision-making in marketing, healthcare, and politics, enhancing information diffusion, and controlling the spread of false information. High-dimensional, noisy, and dynamic data that are typical of real-world networks pose a challenge for conventional influence detection techniques, which are frequently founded on measures of centrality. Their ability to accurately identify key influencers is hindered by these limitations. This study presents a hybrid strategy that combines Optimized Fuzzy Clustering with the SALT (Spatial-Temporal Adaptive Learning Technique) algorithm to address these issues. SALT enables the network to learn and adapt from temporal and spatial patterns, whereas fuzzy clustering takes into account the ambiguous and overlapping nature of social group memberships. Together, these methods make it easier and more accurate for the model to find influential nodes. The proposed method outperforms conventional methods in terms of accuracy, adaptability, and computational performance, as demonstrated by experiments on benchmark datasets including the Twitter and Facebook networks. Fuzzy logic and adaptive learning work well together to handle large amounts of dynamic data, making the method useful for detecting influence in changing social networks.