融合知识交互关系的认知诊断深度模型

认知诊断是基于学习数据挖掘学习者潜在认知状态的一种智能评测技术.当前大多数认知诊断模型将学习任务中的知识视为同等重要,未考虑知识间的交互关系,从而影响诊断的准确性,同时也缺乏可解释性.针对上述问题,文中提出融合知识交互关系的认知诊断深度模型,实现学习者认知状态与知识权重的统一表达.同时,实现基于Choquet积分的理想作答反应计算算法.最后提出模糊测度的深度神经网络,预测学习者的作答表现.大量实验表明,文中模型不仅取得较好的预测结果,还能为预测结果提供知识交互层面的解释,具有一定的优越性.     

基于模糊认知图的情感Agent模型研究

人工情感是人工心理的一个主要研究内容。从研究人工情感出发,提出一种基于模糊认知图的情感Agent建模的方法。模糊认知图模型通过在传统认知图模型中引入模糊测度来量化概念间因果关系的影响程度。Agent的知识由内部组元的状态以及组元之间的关系权值进行描述,用简单数值运算代替了复杂的符号逻辑来实现Agent的智能推理和决策。通过实验表明,该模型设计简单、易于扩展、适用性好。     

基于测度递进的模糊认知图及其应用

模糊认知图(fuzzy cognitive map,FCM)具有简单的推理机制和较强的因果关系表达能力,已得到广泛关注和研究,但FCM对专家经验知识具有较强的依赖性,故而限制了在复杂动态系统建模中的应用.基于此,提出了一种测度递进策略的模糊认知图学习方法.利用线性回归算法,学习得到模糊认知图权重矩阵粗模型;将神经网络的权值调整算法应用于权重矩阵粗模型的细化过程,将该模糊认知图模型应用在股票市场,实现对股票日均值的预测.实验结果表明了该建模方式是有效的.     

模糊粗糙集测度的分析分层建模及应急演练控制应用

针对应急救援演练控制的复杂性和难以量化问题,为实现多人参演系统的有效控制,基于分析分层过程法(analytic hierarchy process,AHP),建立一种模糊粗糙集知识测度的综合建模方法.首先,分析模糊粗糙集各类知识测度相关概念、相互联系和各自特点,通过AHP方法,建立模糊规则的分层度量模型并给出了对比矩阵的构造示例,对模糊规则进行更加精细的度量.其次,在分析应急演练知识构成的基础上,提出预案知识提取和模糊关系粗糙集的构建方法;设计了演练过程控制流程和基于本文知识综合测度方法形成的核心控制流程;通过对规则重要性排序,提高规则判别精度,提供规则选择的手段和一种规则冲突消解方法.最终,通过一个简单案例,验证了本文所提的研究工作的可行性.     

基于信任知识库的概率模糊认知图

模糊认知图较难表示概念间因果关系测度的不确定性、因果联系的时空特性及专家对知识的不确定性．在继承模糊认知图模型优点的前提下，在概念间的因果关系中引入条件概率及信任知识库表示，提出基于信任知识库的概率模糊认知图模型．该模型用条件概率及信任知识库表示因果联系的时空特性、专家对知识及概念间因果关系测度的不确定性，从而将因果关系测度的不确定性、因果联系的时空特性及专家对知识的不确定性有效地融入模糊认知图中，自然扩展了模糊认知图模拟因果关系的能力，较大限度地减少了认知图对现实世界模拟的失真．最后通过实验说明了基于信任知识库的概率模糊认知图模型，具有比FCM更强的模拟能力．     

Hamming-Hausdorff距离下区间直觉模糊知识测度及应用

提出了一种基于改进Hamming-Hausdorff距离的区间直觉模糊知识测度（interval-valued intuitionistic fuzzy knowledge measure,IVIFKM）,并应用于图像阈值分割中,获得了更好的图像分割结果.最新研究成果表明,直觉模糊环境下的知识度量包括两个重要方面,即信息量与信息清晰度.基于这种理解,提出新的区间直觉模糊知识测度公理系统.同时,改进并推广标准Hamming-Hausdorff距离,结合理想解法（technique for order preference by similarity to ideal solution,TOPSIS）,建立新的满足所提公理系统要求的区间直觉模糊知识测度.随后,将所提测度模型应用于图像阈值分割中,并根据区间直觉模糊集自身结构特点,进一步提出一种精炼而高效的像素分类规则及图像区间直觉模糊化算法.最后,利用所提测度模型计算图像的区间直觉模糊知识量,确定最佳分割阈值,实现图像分割.实验结果表明,该基于知识驱动的图像阈值分割方法性能表现稳定、可靠,所生成的二值图具有更加优良的性能指标,明显优于其他同类算法.将知识测度新理论引入图像处理领域,为该理论在其他相关领域的潜在应用提供了实例.     

粗糙集的两种相似性度量

粗糙集理论是一种新的处理模糊和不确定性知识的软计算工具.在近似空间中,首先基于集合的上下近似给出了一种粗糙集间的相似度量方法.然后通过定义一种基于粗糙隶属函数的包含度,给出了另外一种粗糙集间的相似度量方法,并分别研究了这两种相似度量方法的有关性质.最后讨论了这两种相似度量方法之间的关系.     

基于熵的模糊信息测度研究

模糊信息测度(Fuzzy Information Measures,FIM)是度量两个模糊集之间相似性大小的一种量度,在模式识别、机器学习、聚类分析等研究中,起着重要的作用.文中对模糊测度进行了分析,研究了基于熵的模糊信息测度理论:首先,概述了模糊测度理论,指出了其优缺点;其次,基于信息熵理论,研究了模糊熵理论,建立了模糊熵公理化体系,讨论了各种模糊熵,在此基础上,提出了模糊绝对熵测度、模糊相对熵测度等模糊熵测度;最后,基于交互熵理论,建立了模糊交互熵理论,进而提出了模糊交互熵测度.这些测度理论,不仅丰富与发展了 FIM理论,而且为模式识别、机器学习、聚类分析等理论与应用研究提供了新的研究方法.     

基于工作流的知识流建模与控制

知识在多个参与者之间的产生、传播与应用称为知识流.在知识密集型组织中,对业务过程的控制和对知识资产的管理具有紧密的依赖关系.工作流管理是实现业务过程控制的重要技术.当前的工作流过程元模型不支持对知识管理机制的表示.为此,提出了一个扩展的工作流过程元模型,以支持业务过程控制与知识管理的集成.在此基础上,对知识流的建模与控制进行了深入的研究.提出了一种知识流建模方法,通过5类知识流单元对知识传递与重用、人员协作与交流进行表示.针对知识流中的动态因素,研究了基于资源约束、知识需求变化和时间约束的知识流控制方法,以实现自适应的知识流控制,并给出了有关算法.为工作流技术与知识管理技术的有效结合提供了一个有益的途径.     

面向认知协作的知识流分析与研究

为了有效地对认知科学中基于群学习方式的认知协作进行研究,首先探讨了基于本体描述的认知基础环境,对知识驱动的认知环境所涉及的一些基本概念进行了定义．进而,基于本体论思想,对支持认知协作的知识应用集成环境进行了探讨．在此基础上,对知识流关系进行了形式化定义,并利用Markov决策过程理论分析研究了知识应用集成环境下协同认知的过程逻辑．最后,对全文进行了总结并提出了进一步的研究方向．     

Knowledge modeling based on interval-valued fuzzy rough set and similarity inference: prediction of welding distortion

Knowledge-based modeling is a trend in complex system modeling technology. To extract the process knowledge from an information system, an approach of knowledge modeling based on interval-valued fuzzy rough set is presented in this paper, in which attribute reduction is a key to obtain the simplified knowledge model. Through defining dependency and inclusion functions, algorithms for attribute reduction and rule extraction are obtained. The approximation inference plays an important role in the development of the fuzzy system. To improve the inference mechanism, we provide a method of similaritybased inference in an interval-valued fuzzy environment. larity based approximate reasoning, an inference result is Combining the conventional compositional rule of inference with simideduced via rule translation, similarity matching, relation modification, and projection operation. This approach is applied to the problem of predicting welding distortion in marine structures, and the experimental results validate the effectiveness of the proposed methods of knowledge modeling and similarity-based inference.     

Divergence-based cross entropy and uncertainty measures of Atanassov’s intuitionistic fuzzy sets with their application in decision making

The uncertainty measure of Atanassov’s intuitionistic fuzzy sets (AIFSs) is important for information discrimination under intuitionistic fuzzy environment. Although many entropy measures and knowledge measures haven been proposed to depict uncertainty of AIFSs, how to measure the uncertainty of AIFSs is still an open topic. The relation between uncertainty and other measures like entropy measures, fuzziness and intuitionism is not clear. This paper introduces uncertainty measures by using new defined divergence-based cross entropy measure of AIFSs. Axiomatic properties of the developed uncertainty measure are analysis, together with the monotony property of uncertainty degree with respect to fuzziness and intuitionism. To adjust the contribution of fuzzy entropy and intuitionistic entropy on the total uncertainty, the proposed cross entropy and uncertainty measures are parameterized. Numerical examples indicate the effectiveness and agility of the biparametric uncertainty measure in quantifying uncertainty degree. Then we apply the cross entropy and uncertainty measures into an optimal model to determine attribute weights in multi-attribute group decision making (MAGDM) problems. A new method for intuitionistic fuzzy MAGDM problems is proposed to show the efficiency of proposed measures in applications. It is demonstrated by application examples that the proposed measures can get reasonable results coinciding with other existing methods.     

Comparing three methods for modeling the uncertainty in knowledge discovery from area-class soil maps

Knowledge discovery has been demonstrated as an effective approach to extracting knowledge from existing data sources for soil classification and mapping. Soils are spatial entities with fuzzy boundaries. Our study focuses on the uncertainty associated with class assignments when classifying such entities. We first present a framework of knowledge representation for categorizing spatial entities with fuzzy boundaries. Three knowledge discovery methods are discussed next for extracting knowledge from data sources. The methods were designed to maintain information for modeling the uncertainties associated with class assignments when using the extracted knowledge for classification. In a case study of knowledge discovery from an area-class soil map, all three methods were able to extract knowledge embedded in the map to classify soils at accuracies comparable to that of the original map. The methods were also able to capture membership gradations and helped to identify transitional zones and areas of potential problems on the source map when measures of uncertainties were mapped. Among the three methods compared, a fuzzy decision tree approach demonstrated the best performance in modeling the transitions between soil prototypes.     

A novel TODIM-VIKOR approach based on entropy and Jensen–Tsalli divergence measure for picture fuzzy sets in a decision-making problem

The picture fuzzy set (PFS) has grown huge attention in the research area of uncertain information from the last few years. Information measures have been widely studied in various fuzzy environments. Therefore, in this paper, we study the entropy and divergence measures under the picture fuzzy environment. First, the paper introduced a new entropy measure to measure the fuzziness degree associated with a PFS. An example is established to show the capabilities of the proposed entropy measure. Second, the paper defines a new Jensen–Tsalli divergence measure for PFS to evaluate the information of discrimination between two PFS. We also discuss several properties of entropy and divergence measures in detail. Then we present a new method, based on proposed entropy and divergence measure, to determine the objective weights of experts for multicriteria group decision making with picture fuzzy information. The final criteria weights are obtained by combining subjective and objective weights for more reliable weightage of evaluation criteria. By using this comprehensive weight-determination technique, the proposed method can effectively reduce the unreasonable impact of the extreme evaluation data on the evaluation results. Further, a new multi-criteria decision-making approach is developed based on the combining concepts of the TODIM and VIKOR method under the picture fuzzy environment. We used TODIM to obtain the overall dominance degree which considers the bounded rationality of decision makers and VIKOR is used to obtain the compromise ranking of alternatives. Lastly, an application of the proposed integrated model is demonstrated to verify the feasibility and usefulness and the outcomes of the proposed model are compared with the outcomes of the existing approaches to indicate its validity. This integrated method can effectively reduce the distortion of decision information and provide extraordinary evaluation results. The proposed approach is used in detecting the major issues due to which a company is facing such breakdowns.     

模糊软集之间的相似性度量及应用

研究模糊软集的不确定度量问题,给出模糊软集的包含度、相似度公理化定义;基于模糊蕴含算子提出新的模糊软集包含度与相似度度量方法,该方法具有一定的普遍性,在某种程度上提供不同的模糊蕴含算子就可得到不同的包含度与相似度。基于新的相似度度量方法构造了一种决策方法并应用于金融企业流动性检测中。     

模糊属性Petri网建模方法及学习模型研究

定性映射易于表达模糊不确定性知识,但其在表达人类认知思维活动动态特征上存在不足;模糊Petri网比较符合人类思维方式,但相关参数不易获得且其自学习能力存在较大局限性。为此,提出一种模糊属性Petri网(FAPN)形式定义及建模方法。在FAPN结构中构建定性基准参数学习方法,通过定性映射定义4类变迁发生的模糊定性判断规则和相应变迁发生后的结果运算公式,给出FAPN模型的推理算法和学习机制,并模拟系统的动态运行过程。分析结果表明,该方法能有效提高FAPN的学习能力,可适用于以定性判断为特点的诊断系统。     

多粒度云模型的相似性度量

当前存在的云模型相似性度量仅局限于单粒度空间,缺乏多粒度云模型的相似性度量的相关研究.因此,文中首先证明知识距离框架的相关性质,并建立知识距离与信息度量、信息粒度之间的联系,在分层递阶粒结构上得到如下结论:同一粒结构中粒空间的粒度差异正相关于知识距离,通过知识距离可将随粒度连续变化的粒空间映射到一维坐标上.最后,在知识距离框架的基础上提出云模型相似性度量方法.实验验证上述结论在云模型粒空间上成立.     

犹豫模糊信息相似性测度与群一致性测度及群决策应用

针对目前关于犹豫模糊运算与测度的研究中存在的不足,首先给出犹豫模糊熵函数的定义,并将其作为犹豫模糊信息不确定性测度,进而提出犹豫模糊信息特征向量概念,以信息特征向量为出发点对犹豫模糊距离测度和相似性测度展开研究;为优化群决策过程,提出基于完全优先关系的群一致性测度概念并研究其性质;最后,提出基于相似性测度和群一致性测度的群决策方法并结合算例验证所提出方法的有效性.