基于Agent的改进贴近度的多属性评价模型

商务智能中，基于Agent的自动谈判利用Agent的各项人工智能优势，日益受到重视。针对当前对其中的多属性评价量化研究不够的现状，结合犹豫模糊理论，构建基于多属性的方案；将属性分为等级属性和数量属性，采用语言术语集对等级属性进行区间数转换，并给出将数量属性转换成区间数的方法；综合以上研究，引入总体贡献率，对理想贴近度模型进行改进，提出基于总体贡献率的相对贴近度模型，从而提出基于Agent的改进贴近度的多属性评价模型，并对模型步骤进行了归纳和梳理；以北京市碳排放权交易为背景拟定算例和数据，通过计算和对比分析验证了该模型的合理性和有效性。

Multi-attribute Evaluation Model Based on Improved Closeness of Agent

In business intelligence, automatic negotiation based on Agent has been paid more and more attention due to its advantages of artificial intelligence. Aiming at the lack of quantitative research on multi-attribute evaluation, firstly, combined with hesitant fuzzy theory, a scheme based on multi-attribute is constructed; secondly, the attribute is divided into grade attribute and quantitative attribute, and the interval number conversion of grade attribute is carried out by linguistic term set, and the method of converting quantitative attribute into interval number is given; thirdly, the overall contribution rate is introduced to carry out the ideal closeness degree model in this paper, the relative closeness model based on the overall contribution rate is proposed, and the multi-attribute evaluation model based on the improved closeness degree based on Agent is proposed, then the model steps are summarized and sorted; finally, taking Beijing carbon emission trading as the background, the rationality and effectiveness of the model are verified through calculation and comparative analysis.