情感交互进化算法及其在多属性决策应用.pdfVIP

情感交互进化算法及其在多属性决策中的应用 1 2 宿翀 ，李宏光 1. 北京化工大学 信息科学与技术学院, 北京 100029 E-mail: suchong@mail.buct.edu.cn 2. 北京化工大学 信息科学与技术学院, 北京 100029 E-mail: lihg@mail.buct.edu.cn 摘 要: 传统的多属性决策方法大多关注属性权重和属性区间的变化问题，而没有考虑决策过程的进化特性， 并且很少涉及人对决策目标的理解及决策偏好对权重的影响。为此，论文提出了一个人机情感交互的进化算法 并应用于多属性决策问题。基于一类刺激响应机制的人工情感计算方法，在人机交互过程中，不断进化人对多 属性决策问题中各属性权重，形成在交互式进化计算框架下的情感偏好进化的决策策略。测试函数和经济效益 排序实例研究表明了技术方法的有效性。 关键词: 情感计算，进化，交互式进化计算，多属性决策 Affective Interaction based Evolutionary Algorithms with Applications in Multi-attribute Decision-making 1 2 Su Chong , Li Hong-guang 1. School of Information Science and Technology, Beijing University of Chemical Technology, Beijing 100029 E-mail: suchong@mail.buct.edu.cn 2. School of Information Science and Technology, Beijing University of Chemical Technology, Beijing 100029 E-mail: lihg@mail.buct.edu.cn Abstract: Traditional multi-attribute decision-making methods rather focus on attribute weights and attribute interval shifts but rarely take into account of the evolutionary characteristics of decision-making, let alone people’s understanding of the decision objectives as well as impacts on the attribute weights of people’s decision preferences. Motivated by these challenges, affective interaction based evolutionary algorithms are presented, along with applications in multi-attribute decision-making. Therein, taking advantage of stimul