Machine Based Identify Influential Nodes in Complex Networks

Identifying influential nodes in complex networks is critical for optimizing information flow, improving resilience, and enhancing decision-making in various domains. This study proposes a hybrid machine learning (ML) approach that integrates Artificial Neural Networks (ANN) and Decision Trees (DT) to assess node influence by combining multiple centrality measures, such as degree, betweenness, and closeness centrality. Compared to traditional methods, the hybrid approach achieves higher classification accuracy (91%) and precision (88%), demonstrating improved reliability in detecting key nodes. Experimental results confirm that this approach balances different network centrality aspects, providing a more comprehensive evaluation. The proposed method was evaluated using multiple accuracy metrics across different datasets to confirm its robustness and adaptability. The comparative analysis with previous studies validates the effectiveness of this novel methodology in enhancing network analysis