The two universes multi-granularity fuzzy rough set model is an effective tool for handling uncertainty problems between two domains with the help of binary fuzzy relations. This article applies the idea of neighborho...The two universes multi-granularity fuzzy rough set model is an effective tool for handling uncertainty problems between two domains with the help of binary fuzzy relations. This article applies the idea of neighborhood rough sets to two universes multi-granularity fuzzy rough sets, and discusses the two-universes multi-granularity neighborhood fuzzy rough set model. Firstly, the upper and lower approximation operators are defined in the two universes multi-granularity neighborhood fuzzy rough set model. Secondly, the properties of the upper and lower approximation operators are discussed. Finally, the properties of the two universes multi-granularity neighborhood fuzzy rough set model are verified through case studies.展开更多
In granular computing granular structures represent knowledge on universe,in this paper several important granular structures are considered.In a general granular structure the notions of interior point, accumulation ...In granular computing granular structures represent knowledge on universe,in this paper several important granular structures are considered.In a general granular structure the notions of interior point, accumulation point and boundary point etc are proposed,by use of these notions and referring to topological method,the lower and upper approximations of a subset of universe are defined such that they are one kind of generalization of the existing approximations based on some special granular structure.Basic properties of new rough set approximations are investigated.Furthermore,granular structures on universe are characterized by the lower and upper approximation operators.展开更多
Rough set philosophy hinges on the granularity of data, which is used to build all its basic concepts, like approximations, dependencies, reduction etc. Genetic Algorithms provides a general frame to optimize problem ...Rough set philosophy hinges on the granularity of data, which is used to build all its basic concepts, like approximations, dependencies, reduction etc. Genetic Algorithms provides a general frame to optimize problem solution of complex system without depending on the domain of problem.It is robust to many kinds of problems.The paper combines Genetic Algorithms and rough sets theory to compute granular of knowledge through an example of information table. The combination enable us to compute granular of knowledge effectively.It is also useful for computer auto-computing and information processing.展开更多
针对现有的异常检测方法大多无法有效处理不完备混合数据的问题,提出一种面向不完备混合数据的模糊多粒度异常检测算法ADFIIS(Anomaly Detection in Fuzzy Incomplete Information System),所提算法考虑在标称属性和在数值属性上出现缺...针对现有的异常检测方法大多无法有效处理不完备混合数据的问题,提出一种面向不完备混合数据的模糊多粒度异常检测算法ADFIIS(Anomaly Detection in Fuzzy Incomplete Information System),所提算法考虑在标称属性和在数值属性上出现缺失值的情况,能处理混合属性数据。首先,定义属性之间的模糊相似度;其次,计算每个属性的模糊熵,基于熵的大小使用多粒度的思想构建多个属性序列;再次,计算每个样本的异常值以表征它的异常程度;最后,设计相应的ADFIIS算法并分析它的复杂度。在公开数据集上进行实验,将所提算法与ILGNI(Incomplete Local and Global Neighborhood Information network)等主流离群点检测算法对比。实验结果表明,ADFIIS在不完备混合数据集上的受试者操作特征(ROC)曲线效果更好。ADFIIS的曲线下面积(AUC)的平均值优于90%的对比方法,相较于同样能够处理不完备混合数据的ILGNI,它的AUC平均值提升了7个百分点。所提算法使用模型扩展法在不改变原始数据集的情况下对不完备数据集进行异常检测,拓展了异常检测的适用范围。展开更多
为了解决在实际决策时,由于知识背景不同决策者采用不同粒度语言术语集来表达而导致决策结果不准确的问题,本文提出了一种基于多粒度犹豫模糊语言术语集的逼近理想解排序(technique for order preference by similarity to ideal soluti...为了解决在实际决策时,由于知识背景不同决策者采用不同粒度语言术语集来表达而导致决策结果不准确的问题,本文提出了一种基于多粒度犹豫模糊语言术语集的逼近理想解排序(technique for order preference by similarity to ideal solution,TOPSIS)决策方法。首先选用各术语集中的最大粒度作为标准粒度,通过转换算法将每个决策者的语言术语集转换到同一标准粒度下进行集结,得出相应的隶属度语言术语集;然后结合TOPSIS方法,计算每个备选方案与正、负理想点距离,以相对贴近度的大小排序实现最优方案的选择;最后,通过一个实例,验证该方法的可行性和优越性。本文所提方法可应用于最优方案的选择问题中,提升决策结果准确度。展开更多
为同时应对不确定信息表示与风险信息融合对群决策带来的挑战,构建一种三角模糊不完备三支群决策方法,并将其应用于糖尿病诊断决策。首先,针对信息不确定性蕴含的模糊性和不完备性,分别引入三角模糊集和不完备信息系统的概念。通过与多...为同时应对不确定信息表示与风险信息融合对群决策带来的挑战,构建一种三角模糊不完备三支群决策方法,并将其应用于糖尿病诊断决策。首先,针对信息不确定性蕴含的模糊性和不完备性,分别引入三角模糊集和不完备信息系统的概念。通过与多粒度三支决策结合,构建了可调多粒度三角模糊概率粗糙集模型。然后,根据离差最大化法计算属性权重与专家权重,结合ELECTRE(elimination et choice translating reality)方法建立了三角模糊多属性群决策方法。最后,通过对糖尿病患者数据的案例分析和评估,验证了所提方法的可行性和有效性。该方法不仅从不确定信息表示、风险信息融合和最优粒度选择的视角丰富了多粒度三支群决策理论,而且推动了糖尿病智能诊断方面的应用。展开更多
文摘The two universes multi-granularity fuzzy rough set model is an effective tool for handling uncertainty problems between two domains with the help of binary fuzzy relations. This article applies the idea of neighborhood rough sets to two universes multi-granularity fuzzy rough sets, and discusses the two-universes multi-granularity neighborhood fuzzy rough set model. Firstly, the upper and lower approximation operators are defined in the two universes multi-granularity neighborhood fuzzy rough set model. Secondly, the properties of the upper and lower approximation operators are discussed. Finally, the properties of the two universes multi-granularity neighborhood fuzzy rough set model are verified through case studies.
基金supported by grants from the National Natural Science Foundation of China(Nos.11071284 and 61075120)the Natural Science Foundation of Zhejiang Province in China(No.Y107262).
文摘In granular computing granular structures represent knowledge on universe,in this paper several important granular structures are considered.In a general granular structure the notions of interior point, accumulation point and boundary point etc are proposed,by use of these notions and referring to topological method,the lower and upper approximations of a subset of universe are defined such that they are one kind of generalization of the existing approximations based on some special granular structure.Basic properties of new rough set approximations are investigated.Furthermore,granular structures on universe are characterized by the lower and upper approximation operators.
文摘Rough set philosophy hinges on the granularity of data, which is used to build all its basic concepts, like approximations, dependencies, reduction etc. Genetic Algorithms provides a general frame to optimize problem solution of complex system without depending on the domain of problem.It is robust to many kinds of problems.The paper combines Genetic Algorithms and rough sets theory to compute granular of knowledge through an example of information table. The combination enable us to compute granular of knowledge effectively.It is also useful for computer auto-computing and information processing.
文摘针对现有的异常检测方法大多无法有效处理不完备混合数据的问题,提出一种面向不完备混合数据的模糊多粒度异常检测算法ADFIIS(Anomaly Detection in Fuzzy Incomplete Information System),所提算法考虑在标称属性和在数值属性上出现缺失值的情况,能处理混合属性数据。首先,定义属性之间的模糊相似度;其次,计算每个属性的模糊熵,基于熵的大小使用多粒度的思想构建多个属性序列;再次,计算每个样本的异常值以表征它的异常程度;最后,设计相应的ADFIIS算法并分析它的复杂度。在公开数据集上进行实验,将所提算法与ILGNI(Incomplete Local and Global Neighborhood Information network)等主流离群点检测算法对比。实验结果表明,ADFIIS在不完备混合数据集上的受试者操作特征(ROC)曲线效果更好。ADFIIS的曲线下面积(AUC)的平均值优于90%的对比方法,相较于同样能够处理不完备混合数据的ILGNI,它的AUC平均值提升了7个百分点。所提算法使用模型扩展法在不改变原始数据集的情况下对不完备数据集进行异常检测,拓展了异常检测的适用范围。
文摘为了解决在实际决策时,由于知识背景不同决策者采用不同粒度语言术语集来表达而导致决策结果不准确的问题,本文提出了一种基于多粒度犹豫模糊语言术语集的逼近理想解排序(technique for order preference by similarity to ideal solution,TOPSIS)决策方法。首先选用各术语集中的最大粒度作为标准粒度,通过转换算法将每个决策者的语言术语集转换到同一标准粒度下进行集结,得出相应的隶属度语言术语集;然后结合TOPSIS方法,计算每个备选方案与正、负理想点距离,以相对贴近度的大小排序实现最优方案的选择;最后,通过一个实例,验证该方法的可行性和优越性。本文所提方法可应用于最优方案的选择问题中,提升决策结果准确度。
文摘为同时应对不确定信息表示与风险信息融合对群决策带来的挑战,构建一种三角模糊不完备三支群决策方法,并将其应用于糖尿病诊断决策。首先,针对信息不确定性蕴含的模糊性和不完备性,分别引入三角模糊集和不完备信息系统的概念。通过与多粒度三支决策结合,构建了可调多粒度三角模糊概率粗糙集模型。然后,根据离差最大化法计算属性权重与专家权重,结合ELECTRE(elimination et choice translating reality)方法建立了三角模糊多属性群决策方法。最后,通过对糖尿病患者数据的案例分析和评估,验证了所提方法的可行性和有效性。该方法不仅从不确定信息表示、风险信息融合和最优粒度选择的视角丰富了多粒度三支群决策理论,而且推动了糖尿病智能诊断方面的应用。