Nieto-Morote Ana, Ruz-Vila Francisco
Project Engineering Department, Polytechnic University of Cartagena, c/Dr. Fleming, s/n, 30202 Cartagena, Spain.
Department of Electric Engineering, Polytechnic University of Cartagena, c/Dr. Fleming, s/n, 30202 Cartagena, Spain.
Entropy (Basel). 2023 Apr 26;25(5):722. doi: 10.3390/e25050722.
The main objective of this paper is the definition of a membership function assignment procedure based on inherent features of linguistic terms to determine their semantics when they are used for preference modelling. For this purpose, we consider what linguists say about concepts such as language complementarity, the influence of context, or the effects of the use of hedges (modifiers) on adverbs meaning. As a result, specificity, entropy and position in the universe of discourse of the functions assigned to each linguistic term are mainly determined by the intrinsic meaning of the hedges concerned. We uphold that the meaning of weakening hedges is linguistically non-inclusive because their semantics are subordinated to the proximity to the indifference meaning, whereas reinforcement hedges are linguistically inclusive. Consequently, the membership function assignment rules are different: fuzzy relational calculus and the horizon shifting model derived from the Alternative Set Theory are used to handle weakening and reinforcement hedges, respectively. The proposed elicitation method provides for the term set semantics, non-uniform distributions of non-symmetrical triangular fuzzy numbers, depending on the number of terms used and the character of the hedges involved. (This article belongs to the section "Information Theory, Probability and Statistics").
本文的主要目标是定义一种基于语言术语固有特征的隶属函数赋值程序，以便在语言术语用于偏好建模时确定其语义。为此，我们考虑语言学家对诸如语言互补性、语境影响或修饰语（修饰词）对副词意义的影响等概念的论述。结果，分配给每个语言术语的函数在论域中的特异性、熵和位置主要由相关修饰语的内在含义决定。我们坚持认为，弱化修饰语的意义在语言上是不包含的，因为它们的语义从属于与无差异意义的接近程度，而强化修饰语在语言上是包含的。因此，隶属函数赋值规则是不同的：模糊关系演算和源自交替集理论的水平移动模型分别用于处理弱化修饰语和强化修饰语。所提出的启发式方法提供了术语集语义，即非对称三角模糊数的非均匀分布，这取决于所使用的术语数量和所涉及修饰语的性质。（本文属于“信息论、概率与统计”部分）
