A Hybrid Human–AI Consensus Model in Information Granularity-Based Large-Scale Group Decision-Making
Yuhao Wu, Jiawei Zhang
Source abstract
Artificial intelligence (AI) helps people improve decision-making in uncertain environments. When there is an AI decision maker (DM) in a large-scale group, how can consensus be reached? We address this problem by developing a hybrid human–AI consensus model in this paper. First, a two-stage grouping mechanism is proposed: it divides DMs into human and AI at the first stage, then classifies human DMs into several subgroups via the fuzzy c-means (FCM) clustering algorithm based on their fuzzy preference relations (FPRs). Second, a proximity-aware subgroup weight determination method is introduced, considering the cohesion degree, scale, and distance from AI to determine subgroup weights. Then, an information granularity-based model for consensus optimization is proposed, where the objective function is designed to simultaneously minimize the sum of distances from each subgroup’s collective matrix to the global group matrix and maximize the consistency indices of all subgroups. The particle swarm optimization (PSO) algorithm is employed to solve this model, enabling rapid and effective decision-making for large-scale group decision-making (LSGDM) problems in emergency management. A case study is conducted to elucidate the advantages of the developed model, where some comparisons are reported. The results show that human attitudes toward AI play an important role in reaching consensus in LSGDM involving an AI DM.
Evidence graph
No public relationships recorded yet.
Integrity note: This page is a factual metadata record created by deterministic ingestion. It is not a claim that the work moves a mathematical frontier or has been independently verified.