Collaborative decision making (CDM) with linguistic computational techniques has recently achieved significant advancements. Due to the widespread use of sophisticated linguistic constructions, such as generalized comparative linguistic expressions (GCLEs), additional information associated with subjective appraisals has been exploited with the aim of addressing accuracy improvements in multifarious CDM, given that partial information loss is almost inevitable while dealing with complex linguistic comprehension. This paper brings an innovative perspective into CDM from COmponent ANalysis with GCLEs (COANG) to formalize problems involved in making optimal choices, mainly in CDM problems with participants who are usually characterized by domain specificity. Consequently, the focus of this paper is on the domain-specific CDM (DSCDM) in which individual semantics should be built predominantly to model various implications of their decision appraisals with heterogeneity in the knowledgeable domain for the effort of computational reinforcements. The attitude orientation and strength are crucial decompositions to incorporate COANG into DSCDM to establish an elastic paradigm that puts forward individual perception comprehension ahead of exerting collective efforts. The DSCDM based on COANG model enables agents to turn complex challenges of sophisticated linguistic constructions into substantial opportunities by translating them into customized individual semantics (CIS), and CIS into useful insights for making better decisions and improving results. The potential advantages of the proposed COANG-based DSCDM framework are validated with a clinical psychological practice related to the severity assessment of symptoms of schizophrenia.
Read full abstract