On the Term Set's Semantics for Pairwise Comparisons in Fuzzy Linguistic Preference Models.

Area of Science:

• Information Theory, Probability and Statistics
• Linguistics
• Fuzzy Set Theory

Background:

• Linguistic terms and hedges significantly influence preference modeling.
• Understanding the semantics of linguistic terms requires analyzing inherent features and contextual factors.
• Existing methods may not adequately capture the nuanced meanings introduced by hedges.

Purpose of the Study:

• To define a procedure for assigning membership functions to linguistic terms based on their inherent semantic features.
• To determine the semantics of linguistic terms within preference modeling contexts.
• To differentiate and model the distinct semantic impacts of weakening and reinforcement hedges.

Main Methods:

• Analysis of linguistic concepts: language complementarity, context influence, and hedge effects on adverbial meaning.
• Application of fuzzy relational calculus for weakening hedges.
• Utilization of the horizon shifting model from Alternative Set Theory for reinforcement hedges.

Main Results:

• Specificity, entropy, and position of membership functions are determined by hedge semantics.
• Weakening hedges are linguistically non-inclusive; reinforcement hedges are linguistically inclusive.
• The elicitation method yields non-uniform, non-symmetrical triangular fuzzy number distributions.

Conclusions:

• A novel procedure for membership function assignment based on linguistic term features is established.
• Distinct mathematical approaches are necessary to model the differing semantics of weakening and reinforcement hedges.
• The proposed method provides a more accurate representation of term set semantics in preference modeling.