Consensus Process in Large-Scale Group Decision-Making Based on Limited Trust and Social Networks Using Machine Learning

Group decision-making in large-scale environments presents significant challenges due to the complexity of interactions and the necessity for efficient consensus-reaching mechanisms. Traditional approaches often struggle with balancing decision accuracy and computational efficiency, especially in dynamic settings where trust among experts varies. Addressing these challenges, this research presents a novel approach to improving the Consensus Reaching Process (CRP) in Large-Scale Group Decision-Making (LSGDM) based on limited trust and social networks. The primary objective of this method is to enhance the accuracy and speed of the group decision-making process by utilizing machine learning algorithms such as Random Forest in conjunction with social network-based feature learning. The proposed method incorporates various machine learning mechanisms to optimize the consensus process in large-scale group decision-making based on limited trust. Simulation results indicate that the proposed method outperforms traditional approaches, such as Weighted Fuzzy C-Means (WFCM), in reducing the time required to reach a consensus and improving decision-making quality. The evaluation, conducted using PBM and Dunn indices, clearly demonstrates the superiority of the proposed method across most experimental datasets. In addition to numerical improvements, the analysis of trust relationships among experts revealed that this approach has effectively strengthened group interactions and increased expert satisfaction. The practical applications of this method span various fields, including supply chain management, recommender system design, and marketing data analysis, where it can contribute to more efficient decision-making processes. Overall, this research introduces a novel framework for utilizing machine learning in optimizing the group decisionmaking process, which can also be highly effective in areas such as e-commerce and social network management. The results show that the proposed method achieves up to a 63.37% improvement in consensus quality based on the Dunn index and a 33.33% improvement based on the PBM index, demonstrating its effectiveness in large-scale decision-making scenarios.