Managing personalized individual semantic with improved vector expression in multi-granularity linguistic group decision making

Group linguistic assessment with the vector symbolic of linguistic evaluation information has been recently proposed for qualitative group decision making. Due to various individualized characteristics and knowledge levels, evaluators in group assessment often provide linguistic terms based on different individual linguistic evaluation scales to express their preferences on alternatives. In some situations, decision maker needs to distinguish different meanings of the same linguistic term in different individual evaluators’ understandings. To further develop the resolution and the operational performance of the vector expression of linguistic term in multi-granularity linguistic group decision making (MGLGDM), in this study, we present the concept of improved vector expression of linguistic term. First, we present a method of rewriting the numerical symbolic of linguistic term into the improved vector expression based on the individual linguistic evaluation scale. Based on this, we introduce an approach to compare individual linguistic evaluation scales in MGLGDM. Then, an algorithm with improved vector expression is proposed for ranking alternatives in MGLGDM. Finally, a case illustration and some comparative studies have shown that the new proposed algorithm with improved vector expression of linguistic term is accurate and efficient in distinguishing and computing linguistic evaluation information in MGLGDM.