正态云概念的漂移性度量及分析

云模型把自然语言中定性概念的随机性和模糊性有机地综合在一起,通过正向云变换和逆向云变换实现了概念内涵与外延之间的相互转换。基于正态分布和高斯隶属函数的正态云具有普适性。文中通过正态云概念的外包络曲线,根据KL散度刻画分布函数之间差异的特性,定义了一种正态云概念的漂移性度量。最后,结合人类认知的特点,利用该方法以计算的方式对概念在认知过程中可能发生的漂移性进行了模拟研究和实验分析。     

定性规则的云表示

云模型是定性定量间的不确定转换模型，它将概念的模糊性和随机性集成在一起，利用云模型可以有效地表示基于语言值的定性规则。但已有的文献中，定性规则的云表示多种多样，文章根据各种云的内在联系提出了一维正态云的定性规则的统一表示。     

正态云关联规则的预测方法

在信息处理领域,用数据挖掘方法发现关联规则和进行预测是两大热点.文中应用聚类的方法确定正态云的两个参数,并借助正态云模型来划分数量属性的论域,由此生成一系列的正态云关联规则.接着给出了正态云关联规则的挖掘和预测方法.由于用正态云表示的语言值能很好地表达抽象的概念,从而使得挖掘出的正态云关联规则与预测的结果更抽象、更容易被人理解.     

一种新的逆向云算法

云模型是一种可以有效地实现语言值表示的定性概念与其定量表示之间不确定性转换的模型。文章对正态云作了较深入地概率统计分析,指出了现有逆向云算法的不足,并提出了一种新的精确度更高的逆向云算法,最后通过模拟实验比较了新算法与原逆向云算法。     

基于多维正态云模型的控制器研究

在分析一维正态云模型及规则推理的基础上,研究了N维正态云模型及规则推理,及其N维正态云模型规则推理结构,并研究给出了基于N维正态云模型的控制器的设计方法和步骤。通过三维正态云模型控制器的设计表明所给出的N维正态云模型控制器的设计方法是可行的。     

基于云的概念空间模型研究

从认知科学出发,讨论了G?rdenfors的概念空间理论,用云模型对概念空间进行了形式化研究。由于概念、属性中存在着大量的模糊性和不确定性,将云模型和G?rdenfors的概念空间模型结合起来,建立了一套基于云的概念空间模型。并进一步对概念、相似性等进行定义,形成了一套新的基于云的概念空间模型方法。通过一个昆虫分类实例来进行概念空间的建模实验。用静态和动态两种不同的方法来验证基于云模型的概念空间的有效性。     

云模型在测井曲线分层中的应用

针对由模糊信息导致测井曲线人工分层的不准确,提出了一种基于云模型的测井曲线自动分层方法.而云模型能同时体现概念的随机性和模糊性,利用云变换从测井曲线中提取定性概念,然后对提取的概念进行概念跃升,使得算法输出的概念层次更符合实际.实验表明,该方法能够较准确地提取和表征定性概念,所提取的概念中心与传统测井曲线分层方法得到的概念中心吻合较好,验证了算法的有效性,提高了测井曲线分层的准确性.     

云发生器的软件实现*

用MATLAB实现了云发生器,包括正向云发生器和逆向云发生器。正向云发生器的核心技术是用中心极限定理生成正态随机数。对随机数产生的原理进行了阐述,通过用云表示定性概念“青年”对实验结果进行了分析。     

正态云关联规则在预测中的应用

在信息处理领域,用数据挖掘方法发现关联规则和进行预测是两大热点,首先借助正态云模型来替代对数量属性论域的划分,并提出数量属性上的正态云关联规则的概念,接着给出挖掘正态云关联规则的方法,并利用已挖掘出的正态云关联规则进行预测,由于正态云模型较好地软化了数量属性论域的划分边界,从而使得挖掘出的正态云关联规则与预测的结果更容易被人理解。     

基于云模型的不确定性知识表示

云模型是用语言值表示的定性概念与其定量表示之间的不确定性转换模型,论文通过对云模型数学性质的研究,分析云模型的产生背景,证明其表示不确定性知识的合理性和有效性,最后给出了一个应用实例。     

一种基于分形的创新概念设计系统的实现

改进了以往采用数学函数生成初始构件的概念设计过程，提出了一种基于分形的交互式创新概念设计系统的实现方法。给出了一个建筑造型创新设计实例：初始化种群先经分形化的嵌套、拟态等操作再进化计算，使生成的建筑体具有分形自相似的特征。实验表明，分形在创新进化系统中有很好的应用前景。     

基于定性映射模型的概念层次树构建方法

对概念层次树进行简介，分析量-质转化的定性映射（Qualitative Mapping，QM）数学模型，通过对定性基准[αi,βi]的调整，实现概念提升，以构建概念层次树．该方法与人类的思维推理规则较为吻合．     

基于多种群协同进化计算的创新概念设计

针对建筑设计领域应用计算机辅助设计存在的一些问题,通过深入分析概念设计过程中的创新思维和各类创新技法,提出一种基于多种群协同进化的概念创新设计方法,并阐述其具体的实现技术.同时将进化计算应用到创新设计领域,能够取得传统设计方法所无法得到的结果.     

基于非线性云化的自适应帝王蝶优化算法

针对帝王蝶算法多样性退化、易陷入局部最优导致寻优精度不高的问题,提出一种基于非线性云化的自适应帝王蝶算法(NCSMBO).深入探究帝王蝶算法的进化机制,指出其本质为网格式搜索算法;在迁移和调整算子中,采用正向正态云发生器对父代帝王蝶个体执行非线性云化操作,增加候选解的数量,提高局部开发能力;对云化后的后代个体引入贪婪策略,增强算法的可行性;为从发生概率上对突变进行控制,进一步给出双圆正切形式的自适应调整率.在12个不同特征基准测试函数上对包含NCSMBO在内的7种优化算法进行综合评估,以及对两类数学规划问题求解验证,实验验证结果均表明所提算法具有更高的收敛精度和稳定性.     

基于模式识别的创新概念设计

论文提出了一种基于遗传算法的自动求解算法,并将该算法在CAD中的应用作了展望。该算法采用基元模式识别方法识别原有的以图像形式存储的模型,在适应度函数中引入形象度和抽象度的标准并辅以人机交互的方式来判别算法运行过程中产生的个体。实验结果表明,该算法可以充分利用原有设计图案,并能够在运行过程中产生大量的优良个体。     

Competing risk theory approach to the nature of the outcome of conception

An attempt has been made to obtain the probabilities of pregnancy termination by live birth, still birth and abortion, in any particular order, assuming that the termination of pregnancy is associated with three competing ‘risks’, i.e. live birth, still birth and abortion, which need not necessarily operate at the same point of time. A probability model has been constructed to obtain the conditional probability of live birth/still birth/abortion given that the previous pregnancy has been terminated by a particular one of the three. The correlation coefficient between the outcomes of two particular consecutive pregnancy terminations, viz. live birth and reproductive wastage (which includes both abortion and still birth), has also been obtained. Given data pertaining to sample correlation coefficients between the nature of consecutive orders of the outcomes of conception from first to sixth, an exercise has been undertaken to estimate parameters relating to the ‘hazard’ rates of live births, still birth and abortion. The exercise further enables one to test the well-known demographic hypothesis that reproductive wastages follow each other in succession over consecutive orders of the outcomes of conception.     

Nine- to fourteen-year-old children's conception of computers using drawings

This paper deals with one of several methods which were used to investigate the existence, nature and factors influencing a child's (9-14 years) conception of a computer system. Drawings by subjects as a method of data collection is introduced and discussed. Three empirical studies are described which examine the child's model in terms of its components, conduits and causal effect. The factors of age and experience are highlighted as is the discovery that task orientation led to the implication that no single mental model exists.     

Nine- to fourteen-year-old children's conception of computers using drawings

Abstract

This paper deals with one of several methods which were used to investigate the existence, nature and factors influencing a child's (9-14 years) conception of a computer system. Drawings by subjects as a method of data collection is introduced and discussed. Three empirical studies are described which examine the child's model in terms of its components, conduits and causal effect. The factors of age and experience are highlighted as is the discovery that task orientation led to the implication that no single mental model exists.     

From a rule-based conception to dynamic patterns. Analyzing the self-organization of legal systems

The representation of knowledge in the law has basically followed a rule-based logical-symbolic paradigm. This paper aims to show how the modeling of legal knowledge can be re-examined using connectionist models, from the perspective of the theory of the dynamics of unstable systems and chaos. We begin by showing the nature of the paradigm shift from a rule-based approach to one based on dynamic structures and by discussing how this would translate into the field of theory of law. In order to show the full potential of this new approach, we start from an experiment with NEUROLEX, in which a neural network was used to model a corpus of French Council of State decisions. We examine the implications of this experiment, especially those concerning the limits of the model used, and show that other connectionist models might correspond more adequately to the nature of legal knowledge. Finally, we propose another neural model which could show not only the rules which emerge from legal qualification (NEUROLEX's goal), but also the way in which a legal qualification process evolves from one concept to another.     

Three problems with the connectivist conception of learning

Connectivism, which has been argued to be a new learning theory, has emerged in the field of online learning during the last decade. On the World Wide Web at least, connectivism promises to establish learning spaces similar to those that Ivan Illich imagined in ‘Deschooling Society’, through so‐called massive online open courses (MOOCs). In this paper, we critically examine the theoretical postulates of connectivism and identify three important psychological and epistemological problems, namely the lack of a solution to the learning paradox, the underconceptualization of interaction and the inability to explain concept development. Some of the theoretical deficiencies in question may explain certain learning problems experienced by participants in MOOCs. The paper concludes that, although MOOCs are a worthwhile experience and ought to be continued, connectivism as a learning theory has significant theoretical problems and should be profoundly revised if it is to explain and foster learning in such environments.