摘要
A semantics-based model is proposed to enable weakened hedges, such as ＂more or less＂ and ＂roughly＂ in the context of linguistic multi-criteria decision making. First, the resemblance relations are defined based on the semantics of terms on the domain. Then, the hedges can be represented after the upper and loose upper approximations of a linguistic term are derived. Accordingly, some compact formulae can be derived for the semantics of linguistic expressions with hedges. Parameters in these formulae are objectively determined according to the semantics of original terms. The proposed model presents a more natural way to express the decision information under uncertainties and its semantics is clear. The proposed model is clarified by solving the problem of evaluation and selection of sustainable innovative energy technologies. Computational results demonstrate that the model can deal with various uncertainties of the problem. Finally, the model is compared with existing techniques and extended to the case when the semantics of terms are represented by trapezoidal fuzzy numbers.
为允许在语言多属性决策环境中使用"差不多"和"有点"等表示程度弱化的语言修饰词,提出了一种基于语义的表示模型.首先,基于论域上语言标度的语义定义相似关系,并通过计算语言标度的上近似和松上近似得到语言修饰词的表示方法,进而推导出语言表达式的语义公式,其中参数由语言标度的语义客观确定.该模型具有清晰的语义,且允许专家以更自然的方式表达不确定的决策信息.将模型应用于可持续创新能源技术的评估和选择,计算结果表明,所提出的模型可以处理评估过程中存在的多种不确定性.最后,讨论了该模型与现有技术的区别,并给出了该模型在语义为梯形模糊数情形下的拓展.
基金
The National Natural Science Foundation of China(No.61273209)
the Scientific Research Foundation of Graduate School of Southeast University(No.YBJJ1528)
the Scientific Innovation Research of College Graduates in Jiangsu Province(No.KYLX15-0191)
