Group Decision Method Without Consensus Threshold Based on Personalized Semantic Continuous Learning

The primary concern in group decision-making lies in the objective reasonable creation of a decision outcome that is agreed upon by all decision-makers. To achieve this purpose, a dynamic adjustment of preferences is necessary, in which personalized semantic continuous learning of linguistic information initially provided by decision-makers is a key process. This study explores a way for guiding decision-makers to continuously learn from individual linguistic preferences and establish an adaptive consensus-reaching method. The continuous personalized semantic learning model is firstly designed to simulate individual preferences in a dynamic decision-making environment, addressing the issue of quantifying semantics for decision-makers. Secondly, an adaptive weight allocation method is proposed to capture the changing process of a decision maker's weight while measuring its importance based on the current decision environment. Furthermore, we establish an adaptive consensus-reaching model without subjective parameters facilitates objective evaluation of decision-making. Finally, some experiments are conducted to examine the effectiveness of the proposed model.