A Personalized Feedback Mechanism Based on Bounded Confidence Learning to Support Consensus Reaching in Group Decision Making

Different feedback mechanisms have been reported in consensus reaching models to provide advices for preference adjustment to assist decision makers to improve their consensus levels. However, most feedback mechanisms do not consider the willingness of decision makers to accept these advices. In the opinion dynamics discipline, the bounded confidence model justifies well that in the process of interaction a decision maker only considers the preferences that do not exceed a certain confidence level compared to his own preference. Inspired by this idea, this article proposes a new consensus reaching model with personalized feedback mechanism to help decision makers with bounded confidences in achieving consensus. Specifically, the personalized feedback mechanism produces more acceptable advices in the two cases where bounded confidences are known or unknown, and the unknown ones are estimated by a learning algorithm. Finally, numerical example and simulation analysis are presented to explore the effectiveness of the proposed model in reaching consensus.

Automated summary

This article introduces a new consensus reaching model with personalized feedback mechanism to help decision makers with bounded confidences in achieving consensus. The personalized feedback mechanism provides more acceptable advices and can estimate unknown bounded confidences through a learning algorithm.