基于粗糙集的个性化学习平台的研究

现代教育技术在网络和多媒体的推动下飞速发展,但网络中的数据冗余,内容层次不一,这些学习平台提供的服务大都没有考虑到学生个性化的需求。我们通过粗糙集算法对学生在线学习的访问数据将学生进行聚类,从而个性化匹配适合他的学习资源和服务,提高他的学习效能。     

基于粗糙集理论和BP神经网络的数据挖掘算法

根据数据挖掘中粗糙集理论和BP神经网络各自的优势和存在的问题,提出了一种将粗糙集理论和BP神经网络理论结合在一起的算法。该算法利用粗糙集对属性的归约功能将数据仓库中的数据进行归约,并将归约后的数据作为训练数据提供给BP神经网络。通过粗糙集归约,提高了训练数据表达的清晰度,也减小了BP神经网络的规模,同时利用BP神经网络又克服了粗糙集对噪声数据敏感的影响。文中提出了代价函数,解决了训练数据与网络精度的问题,也提供了由粗糙集归约向BP神经网络训练转变的依据。     

Knowledge acquisition in incomplete fuzzy information systems via the rough set approach

Abstract: Machine learning can extract desired knowledge from training examples and ease the development bottleneck in building expert systems. Most learning approaches derive rules from complete and incomplete data sets. If attribute values are known as possibility distributions on the domain of the attributes, the system is called an incomplete fuzzy information system. Learning from incomplete fuzzy data sets is usually more difficult than learning from complete data sets and incomplete data sets. In this paper, we deal with the problem of producing a set of certain and possible rules from incomplete fuzzy data sets based on rough sets. The notions of lower and upper generalized fuzzy rough approximations are introduced. By using the fuzzy rough upper approximation operator, we transform each fuzzy subset of the domain of every attribute in an incomplete fuzzy information system into a fuzzy subset of the universe, from which fuzzy similarity neighbourhoods of objects in the system are derived. The fuzzy lower and upper approximations for any subset of the universe are then calculated and the knowledge hidden in the information system is unravelled and expressed in the form of decision rules.     

Rough Neural Network of Variable Precision

In this paper, a new method is described to construct rough neural networks. On the base of rough set model, we present a method to develop rough neural network of variable precision and train it using Levenberg–Marquart algorithm. The method is particularly attractive because it combines the advantages of both rough logic networks and neural networks. In our system, weak generalization in rough sets theory and complexity in neural network are avoided while anti-jamming performance is highly improved and the network structure is also simplified. In experiments, the network is applied to classification of remote sensing images. The results show that our method is more effective and successful than application of rough sets and neural network separately.     

基于粗糙神经网络的医学图像分类新方法

由于乳腺X光图像的复杂性,直接从图像中看出肿瘤及其良、恶性质是比较困难的,因此建立高效的肿瘤自动诊断系统是非常必要的。文章将粗糙集理论中基于信息增益的约简方法和神经网络相结合,提出了粗糙神经网络算法RNN,将其应用于乳腺X光图像分类。实验结果表明,该方法的分类精确度可达到92.37%比单独使用神经网络方法的分类精确度(81.25%)要高,同时所花费的时间也明显减少。     

基于粗糙集联系度的数据挖掘算法及应用研究

粗糙集理论和模糊集理论都是用来处理不完整和不确定信息的理论，两者都可用来观察、测试数据并进行推理。将集对分析中的联系度概念应用于粗糙集中，说明了粗糙集联系度与下近似集和上近似集的值化的关系。文中分析了专家系统中规则抽取中存在的困难，用粗糙集理论和集对分析理论解决专家系统中规则的抽取和过滤问题，提出了一种新的规则提取方法，并给出了一个应用实例。     

A ROUGH SET/FUZZY LOGIC BASED DECISION MAKING SYSTEM FOR MEDICAL APPLICATIONS

Abstract

A method of performing prognostic modeling of disease states is proposed. The technique uses rough sets to extract rules from a database. The data is then reformatted into a fuzzy logic template, and a learning algorithm is used to adjust the fuzzy set membership functions. The method is applied to the POSCH problem, which looks at risk factors associated with the progression of coronary artery disease. The POSCH data has several shortcomings, including a limited number of cases, correlated inputs, as well as noise on both the inputs and outcome. The problem was to predict progression of atherosclerosis in the LAD three years after baseline based on physiologic data available at baseline. The proposed rough/fuzzy set method correctly predicted progression of atherosclerotic disease in 69% of the patients, which is statistically better than neural network, rough set and logistic models performed.     

基于粗糙集理论的神经网络研究及应用

为了补偿神经网络的黑箱特性并提高其工作性能,将粗糙集理论同神经网络结合起来,提出一种基于粗糙集的神经网络体系结构．首先,利用粗糙集理论对神经网络初始化参数的选择和确定进行指导,赋予各参数相关的物理意义;然后,以系统输出误差最小化为目标对粗糙神经网络进行训练,使其满足性能要求．实验结果表明,粗糙神经网络能较好地完成数据挖掘任务,并能获得较高的分类精度．     

基于RSPNN的制粉系统故障诊断

针对发电厂制粉系统故障与征兆对应关系复杂及过程信息的不确定性及传统BP神经网络故障诊断的缺点,提出了基于粗糙集概率神经网络(RSPNN)的制粉系统故障诊断方法,以改善传统BP神经网络初始值敏感、易使学习过程陷入局部极小值以及样本数据过大时训练速度慢等问题。首先采用自组织映射神经网络(SOMNN)对连续样本数据进行离散化;再利用基于区分矩阵的HORAFA算法对离散化样本数据进行RS属性约简,并将约简结果作为概率神经网络(PNN)的输入;最后利用PNN作为诊断决策分类器,输出故障模式,并进行了仿真研究。仿真结果表明,该方法不仅优化神经网络的拓扑结构,降低神经网络的训练时间,而且能准确、快速地诊断制粉系统故障类型,同时对发电厂制粉系统及其相关设备的在线故障诊断问题有一定启发性。     

粗糙集学习机器泛化性能控制的结构风险最小化方法

对影响粗糙集学习机器泛化性能的因素进行了分析,通过将结构风险最小化原则引入到粗糙集学习中,提出了粗糙集学习的结构风险最小化方法;通过12个UCI数据集上的实验分析,验证了提出方法的有效性.     

基于一种新的核函数的模糊粗糙集

模糊粗糙集作为模糊集与粗糙集的结合体,能够有效处理数据的复杂性和不确定性。由模糊相似关系产生的模糊粒结构可以对模糊粗糙集中不确定性的概念进行近似。核函数和模糊相似关系分别是机器学习和模糊粗糙集的核心因素,因此借助模糊相似关系和核函数之间的关系,构造了一种新的核函数,并定义了相应的核模糊粗糙集。最后通过实例说明新构造的核函数具有一定的推广性。     

基于粗糙集的神经网络建模方法研究

提出了一种基于粗糙集的神经网络模型,该方法利用粗糙集数据分析方法,从数据中 提取出规则将输入映射到输出的子空间上,而后在这个子空间上用神经网络进行逼近.利用这 种方法对岩石边坡工程中边坡稳定性进行分析建模,并和传统的神经网络建模方法进行比较, 说明了该方法的有效性.     

粗糙集约简的神经网络遥感分类应用

近年来,随着人工神经网络系统理论的发展,神经网络技术日益成为遥感数字图像分类处理的有效手段。但是该方法不能降低维数、时间开销大,针对这些不足提出一种基于粗糙集约简的神经网络方法。本文对RapidEye影像进行分析并提取纹理特征,利用粗糙集理论对纹理特征与光谱特征属性进行约简,得到的约简属性作为输入属性,利用神经网络法对影像分类。结果表明该方法具有较好的分类精度。     

基于粗糙集理论的网络型入侵检测系统

为解决目前大多数入侵检测产品或模型对未知攻击的检测都存在精度低或者虚警率高的问题，建立了一个基于网络的入侵检测实验平台，使用了多种新的攻击工具实施攻击；并在此基础上提取了网络连接的29项实时特征；应用粗糙集理论实现了一个网络连接的检测器。经实验表明，所选取的网络连接特征能较好地反映网络安全状况，粗糙集理论应用于多类分类问题和未知攻击的检测方面是有效的。     

基于粗糙集理论的模式分类样本特征选择方法研究

本文提出了一种基于粗糙集理论的模式分类本特征选择方法，该方法应用粗糙集理论和方法，对给定的学习样本进行特征选择，根据这些特征构造神经网络模型进行训练，并在网络的工作阶段，根据这些特征对待识样本进行分类，在模式分类中，该方法能够减少网络的训练时间并改善网络的泛化能力。     

Consistency and Completeness in Rough Sets

Consistency and completeness are defined in the context of rough set theory and shown to be related to the lower approximation and upper approximation, respectively. A member of a composed set (union of elementary sets) that is consistent with respect to a concept, surely belongs to the concept. An element that is not a member of a composed set that is complete with respect to a concept, surely does not belong to the concept. A consistent rule and a complete rule are useful in addition to any other rules learnt to describe a concept. When an element satisfies the consistent rule, it surely belongs to the concept, and when it does not satisfy the complete rule, it surely does not belong to the concept. In other cases, the other learnt rules are used. The results in the finite universe are extended to the infinite universe, thus introducing a rough set model for the learning from examples paradigm. The results in this paper have application in knowledge discovery or learning from database environments that are inconsistent, but at the same time demand accurate and definite knowledge. This study of consistency and completeness in rough sets also lays the foundation for related work at the intersection of rough set theory and inductive logic programming.     

Rough Neural Computing in Signal Analysis

This paper introduces an application of a particular form of rough neural computing in signal analysis. The form of rough neural network used in this study is based on rough sets, rough membership functions, and decision rules. Two forms of neurons are found in such a network: rough membership function neurons and decider neurons. Each rough membership function neuron constructs upper and lower approximation equivalence classes in response to input signals as an aid to classifying inputs. In this paper, the output of a rough membership function neuron results from the computation performed by a rough membership function in determining degree of overlap between an upper approximation set representing approximate knowledge about inputs and a set of measurements representing certain knowledge about a particular class of objects. Decider neurons implement granules derived from decision rules extracted from data sets using rough set theory. A decider neuron instantiates approximate reasoning in assessing rough membership function values gleaned from input data. An introduction to the basic concepts underlying rough membership neural networks is briefly given. An application of rough neural computing in classifying the power system faults is considered.     

一种基于FRS-FCM算法的集成入侵检测方法的研究

传统FCM算法对初值的依赖性过大且欧氏距离只适用于处理数值型及特征空间为超球结构的数据集。为此,利用模糊粗糙集思想,结合ReliefF技术,提出了一种基于模糊粗糙集的特征加权聚类算法(FRS-FCM),并将此算法应用到集成入侵检测中,通过有效地聚类和集成学习来提高入侵检测的检测率,降低误检率,并较大地提高低频攻击的检测率。最后利用KDD Cup 99数据集进行的仿真实验验证了该方法的可行性与有效性。     

基于粗糙集神经网络的网络故障诊断新方法

摘要针对传统网络故障知识库冗余度高和稳定性难以两全的缺陷，综合运用神经网络方法和粗糙集理论，提出了RSNN算法，实现不一致情况下的规则获取和学习样本的净化处理．该算法具有简化样本、适应性强、容错性高和不易陷入局部最小点等特点，能有效处理网络故障诊断中噪声或不相容的信息．实验表明，利用该方法实现的系统与同类的其他方法相比，提高了诊断准确率和诊断速度．