Punishment-driven consensus reaching model in social network large-scale decision-making with application to social capital selection

Su-min Yu,Zhi-jiao Du,Xue-yang Zhang,Han-yang Luo,Xu-dong Lin
DOI: https://doi.org/10.1016/j.asoc.2021.107912
IF: 8.7
2021-12-01
Applied Soft Computing
Abstract:Clustering and consensus reaching are two crucial processes for solving social network large-scale decision-making (SN-LSDM) problems. This paper explores the moderating effect of trust on the decision-making process, which is decomposed into examining the influence of trust constraint on clustering and the influence of trust loss on consensus reaching. We hold that low-trust experts are not suitable to be assigned to the same subgroup, even though their preferences are sufficiently similar. We propose a punishment-driven consensus reaching model for SN-LSDM problems. The model contains two main stages. In Stage 1, a trust-constrained K-means clustering algorithm is presented, which can overcome the defect of grouping low-trust experts caused by traditional clustering methods based on preference similarity. Then, a weight-determining method combining the size, cohesion, and overall trust degree of a subgroup is proposed. In Stage 2, a punishment-driven consensus reaching process is designed, which identifies four categories of consensus scenarios (high–high, high–low, low–high, low–low) through consensus measures and trust measures. Different adjustment strategies are established, and the moderating effect of trust loss on consensus reaching and consensus cost is investigated. Finally, both a case study on social capital selection in public–private partnership projects and a comparative analysis are implemented to reveal the feasibility and advantages of the clustering algorithm and consensus reaching model proposed in this paper.
computer science, artificial intelligence, interdisciplinary applications
What problem does this paper attempt to address?