Integration of Improved Whale Optimization Algorithm and Deep Learning for Trust Management in Social IoT Environment

The Social Internet of Things (SIoT) integrates IoT with social networks, enabling autonomous device interactions and seamless data sharing.A key challenge in SIoT is ensuring secure and reliable interactions through an effective Trust Management System (TMS).This paper proposes the Integration of Improved Whale Optimization Algorithm and Deep Learning for Trust Management System (IIWOADL-TMS) to enhance trust evaluation and authentication in SIoT environments.The model computes direct and indirect trust values to assess device credibility, restricting access to untrusted nodes.It employs Long Short-Term Memory (LSTM) networks for traffic classification and dynamically determines the Threshold Trust Value (THVL) to enhance security.The Improved Whale Optimization Algorithm (IWOA) optimizes LSTM's hyperparameters, improving classification accuracy and trust assessment efficiency.Experimental evaluations demonstrate that IIWOADL-TMS outperforms existing models in trust evaluation, authentication, and secure data communication.The proposed system enhances reliability, scalability, and resilience in SIoT networks by mitigating security risks and improving trust computation mechanisms.These results highlight IIWOADL-TMS as a robust solution for secure and efficient interactions in SIoT.