Detection of Influential nodes using Hybrid Deep learning methods in IIOT environment

Abstract Influential nodes in an Industrial Internet of Things (IIoT) environment using Beluga Whale Optimization Algorithm (BWO) integrated with Residual Long Short-Term Memory (LSTM) networks. In IIoT networks, identifying influential nodes is crucial for optimizing data transformation, minimizing latency, and improving overall network effectiveness. The proposed method leverages the exploration and exploitation capabilities of the Beluga Whale Optimization (BWO) Algorithm to optimize the parameters of an Recurrent Long Short-Term Memory (RLSTM) model, which is used to predict the behavior of nodes and identify key influencers within the network. The integration of BWO with RLSTM helps improve the accuracy of node predictions by dynamically adjusting the RLSTM’s hyperparameters based on the network’s evolving data. Extensive experiments conducted in a simulated IIoT environment highlight the performance of the proposed model in enhancing prediction accuracy, reducing computational overhead, and improving network efficiency compared to traditional methods. The results highlight the potential of this hybrid optimization technique for real-time applications in smart manufacturing, predictive maintenance, and other IIoT-driven sectors.