Two-Fold Personalized Feedback Mechanism for Social Network Consensus by Uninorm Interval Trust Propagation

Abstract

A twofold personalized feedback mechanism is established for consensus reaching in social network group decisionmaking (SN-GDM). It consists of two stages: (1) generating the trusted recommendation advice for individuals; and (2) producing personalized adoption coefficient for reducing unnecessary adjustment costs. This is achieved by means of a uninorm interval-valued trust propagation operator to obtain indirect trust. The trust relationship is used to generate personalized recommendation advice based on the principle of ‘a recommendation being more acceptable the higher the level of trust it derives from’. An optimization model is built to minimise the total adjustment cost of reaching consensus by determining personalized feedback adoption coefficient based on individuals’ consensus levels. Consequently, the proposed twofold personalized feedback mechanism achieves a balance between group consensus and individual personality. An example to demonstrate how the proposed twofold personalized feedback mechanism works is included, which is also used to show its rationality by comparison with the traditional feedback mechanism in GDM.