决策粗糙集均值模型【毕业论文】.doc

本科毕业论文 （20 届） 决策粗糙集均值模型 专业：数学与应用数学 摘要 网络信息时代知识急剧增加，面对大量的、杂乱无章的数据，人们希望能从中挖掘出潜在的、有用的信息。粗糙集理论与方法对于处理复杂的系统，无需提供已知集合之外的任何先验信息，不失为一种有效的处理方法。但Pawlak粗糙集模型是基于完全确定的等价关系的，它忽视了不确定知识的可利用信息。这类模型在对随机产生的知识库的数据分析方面往往不能完全反映问题的实质。因此，许多学者为研究不确定性系统提出了新的粗糙集模型——概率粗糙集模型。本文在对经典粗糙集、概率粗糙集进行分析的基础上，针对当多用户参与决策时，在每个用户做出判断的前提下，为了综合每个用户的意见，提出基于均值的决策模型 A Mean Value Decision-theoretic Rough Set Model Abstract As knowledge increased dramatically in the network information age, facing to massive, chaotic data, people want to dig out the potential and useful information. For dealing with complex systems, rough set theory and methods are not necessary to be provided with any prior information other than the collection of the required processing, which is an effective approach. But the Pawlak rough set model is based on the equivalence relation, which ignores the available information of uncertain .For the data analysis of the randomly generated knowledge data, these models often does not fully reflect the essence of the problem. Therefore, in order to study the uncertain system, many scholars proposed a new rough set probabilistic rough set model. Based on the analysis of the classical rough sets and the probabilistic rough sets, the multiple agents participate in decision-making. In this model, each user makes judgments. In order to synthesize the views of agents, the decision-theoretic model based on mean value is proposed in this thesis and the best decision to the application is given which is demonstrated to be its usefulness in life. Keywords: Rough set; Decision-theoretic rough set; Multi-agents decision-making; Mean value 目录 摘要 I Abstract II 1．前言 1 2．Pawlak代数粗糙集模型 3 3．概率粗糙集模型 4 4．决策粗糙集模型 5 4.1单用户决策 5 4.2多用户决策 7 5.多用户决策粗糙集模型实例 9 结论 12 参考文献 13 致谢 15  1．[14-17]提出了在Bayes分析基础上的决策粗糙集模型（decision-theoretic rough sets）；Polkowski[9]和Skowron[11]对概率近似分类与模糊集两者作了比较研究；Pawlak，Wong和Ziarko等人[5]提出了0.5-概率粗糙集模型；Ziarko[20]提出了可变精度粗糙集模型；Pawlak和Skowron等人[8]引入了粗糙隶属度和粗糙隶属函数的概念；Skowron和Stepaniuk提出了参数化粗糙集模型；Greco