Attribute Reduction of Hybrid Decision Information Systems Based on Fuzzy Conditional Information Entropy

The presence of numerous uncertainties in hybrid decision information systems(HDISs)renders attribute reduction a formidable task.Currently available attribute reduction algorithms,including those based on Pawlak attribute importance,Skowron discernibility matrix,and information entropy,struggle to effectively manages multiple uncertainties simultaneously in HDISs like the precise measurement of disparities between nominal attribute values,and attributes with fuzzy boundaries and abnormal values.In order to address the aforementioned issues,this paper delves into the study of attribute reduction withinHDISs.First of all,a novel metric based on the decision attribute is introduced to solve the problem of accurately measuring the differences between nominal attribute values.The newly introduced distance metric has been christened the supervised distance that can effectively quantify the differences between the nominal attribute values.Then,based on the newly developed metric,a novel fuzzy relationship is defined from the perspective of“feedback on parity of attribute values to attribute sets”.This new fuzzy relationship serves as a valuable tool in addressing the challenges posed by abnormal attribute values.Furthermore,leveraging the newly introduced fuzzy relationship,the fuzzy conditional information entropy is defined as a solution to the challenges posed by fuzzy attributes.It effectively quantifies the uncertainty associated with fuzzy attribute values,thereby providing a robust framework for handling fuzzy information in hybrid information systems.Finally,an algorithm for attribute reduction utilizing the fuzzy conditional information entropy is presented.The experimental results on 12 datasets show that the average reduction rate of our algorithm reaches 84.04%,and the classification accuracy is improved by 3.91%compared to the original dataset,and by an average of 11.25%compared to the other 9 state-of-the-art reduction algorithms.The comprehensive analysis of these research results clearly indicates that our algorithm is highly effective in managing the intricate uncertainties inherent in hybrid data.

Attribute reduction based on background knowledge and its application in classification of astronomical spectra data

To improve the efficiency of the attribute reduction, we present an attribute reduction algorithm based on background knowledge and information entropy by making use of background knowledge from research fields. Under the condition of known background knowledge, the algorithm can not only greatly improve the efficiency of attribute reduction, but also avoid the defection of information entropy partial to attribute with much value. The experimental result verifies that the algorithm is effective. In the end, the algorithm produces better results when applied in the classification of the star spectra data.

Extension of information entropy-based measures in incomplete information systems

It is helpful for people to understand the essence of rough set theory to study the concepts and operations of rough set theory from its information view. In this paper we address knowledge expression and knowledge reduction in incomplete infolvnation systems from the information view of rough set theory. First, by extending information entropy-based measures in complete information systems, two new measures of incomplete entropy and incomplete conditional entropy are presented for incomplete information systems. And then, based on these measures the problem of knowledge reduction in incomplete information systems is analyzed and the reduct definitions in incomplete information system and incomplete decision table are proposed respectively. Finally, the reduct definitions based on incomplete entropy and the reduct definitions based on similarity relation are compared. Two equivalent relationships between them are proved by theorems and an in equivalent relationship between them is illustrated by an example. The work of this paper extends the research of rough set theory from information view to incomplete information systems and establishes the theoretical basis for seeking efficient algorithm of knowledge acquisition in incomplete information systems.

Fusing Supervised and Unsupervised Measures for Attribute Reduction

It is well-known that attribute reduction is a crucial action of rough set.The significant characteristic of attribute reduction is that it can reduce the dimensions of data with clear semantic explanations.Normally,the learning performance of attributes in derived reduct is much more crucial.Since related measures of rough set dominate the whole process of identifying qualified attributes and deriving reduct,those measures may have a direct impact on the performance of selected attributes in reduct.However,most previous researches about attribute reduction take measures related to either supervised perspective or unsupervised perspective,which are insufficient to identify attributes with superior learning performance,such as stability and accuracy.In order to improve the classification stability and classification accuracy of reduct,in this paper,a novel measure is proposed based on the fusion of supervised and unsupervised perspectives:(1)in terms of supervised perspective,approximation quality is helpful in quantitatively characterizing the relationship between attributes and labels;(2)in terms of unsupervised perspective,conditional entropy is helpful in quantitatively describing the internal structure of data itself.In order to prove the effectiveness of the proposed measure,18 University of CaliforniaIrvine(UCI)datasets and 2 Yale face datasets have been employed in the comparative experiments.Finally,the experimental results show that the proposed measure does well in selecting attributes which can provide distinguished classification stabilities and classification accuracies.

Half-global discretization algorithm based on rough set theory

It is being widely studied how to extract knowledge from a decision table based on rough set theory. The novel problem is how to discretize a decision table having continuous attribute. In order to obtain more reasonable discretization results, a discretization algorithm is proposed, which arranges half-global discretization based on the correlational coefficient of each continuous attribute while considering the uniqueness of rough set theory. When choosing heuristic information, stability is combined with rough entropy. In terms of stability, the possibility of classifying objects belonging to certain sub-interval of a given attribute into neighbor sub-intervals is minimized. By doing this, rational discrete intervals can be determined. Rough entropy is employed to decide the optimal cut-points while guaranteeing the consistency of the decision table after discretization. Thought of this algorithm is elaborated through Iris data and then some experiments by comparing outcomes of four discritized datasets are also given, which are calculated by the proposed algorithm and four other typical algorithras for discritization respectively. After that, classification rules are deduced and summarized through rough set based classifiers. Results show that the proposed discretization algorithm is able to generate optimal classification accuracy while minimizing the number of discrete intervals. It displays superiority especially when dealing with a decision table having a large attribute number.

Attribute reduction in interval-valued information systems based on information entropies

Interval-valued data appear as a way to represent the uncertainty affecting the observed values. Dealing with interval-valued information systems is helpful to generalize the applications of rough set theory. Attribute reduction is a key issue in analysis of interval-valued data. Existing attribute reduction methods for single-valued data are unsuitable for interval-valued data. So far, there have been few studies on attribute reduction methods for interval-valued data. In this paper, we propose a framework for attribute reduction in interval-valued data from the viewpoint of information theory. Some information theory concepts, including entropy, conditional entropy, and joint entropy, are given in interval-valued information systems. Based on these concepts, we provide an information theory view for attribute reduction in interval-valued information systems. Consequently, attribute reduction algorithms are proposed. Experiments show that the proposed framework is effective for attribute reduction in interval-valued information systems.

基于多核模糊条件熵的多类型混合数据属性约简算法

对数据进行有效属性约简是数据挖掘中一个具有挑战性的任务。当前,粗糙集理论是构造属性约简的一种常用方法。然而,现有的属性约简方法都侧重于单类型的数据,对现实环境下多类型混合的数据并不适用。为了解决这一问题,提出一种多核模糊条件熵的多类型混合数据属性约简算法。首先,针对标记型、数值型、区间型和集值型混合的多类型数据,提出了一种多核模糊相似关系。然后,基于这种多核模糊相似关系,定义了一种多核模糊条件熵模型,并讨论了它的单调性和有界性。最后,利用多核模糊条件熵的单调性提出了一种多类型混合数据的属性约简算法。通过UCI数据集的实验分析验证了该算法的有效性。

一种基于信息熵加权的属性约简算法

针对现有邻域粗糙集模型中存在属性权重都相同,无法保证关键属性在属性约简时能够被保留的问题,提出了一种基于信息熵加权的属性约简算法。首先,采用了类间熵、类内熵策略,以最大化类间熵最小化类内熵为原则给属性赋予权重;其次,构造了基于加权邻域关系的加权邻域粗糙集模型;最后,基于依赖关系评估属性子集的重要性,从而实现属性约简。在基于UCI数据集上与其他三种属性约简算法进行对比实验,结果表明,该算法能够有效去除冗余,提高分类精度。

基于自信息熵的直觉模糊决策系统的属性约简

针对在直觉模糊集中,利用下近似构建的约简只考虑了下近似而忽略了上近似,从而导致一些信息丢失的问题,基于直觉模糊集的上、下近似提出了3种熵度量,并将其应用于直觉模糊决策信息系统的约简之中。在直觉模糊决策信息系统上定义用于描述直觉模糊关系的3种不确定性度量,分别为平均决策指数、平均安全决策指数以及平均风险决策指数,并在此基础上依次提出了条件信息熵、条件粗糙熵和自信息熵,基于自信息熵给出了相应的约简定义以及属性约简算法。在多个数据集上的实验表明,所提出的属性约简算法与其他算法相比,约简结果更具有优越性以及鲁棒性。

基于不完备混合序信息系统的增量式属性约简

由于大数据环境下数据呈现出动态更新的特征,因此,增量式属性约简已成为粗糙集理论的重点研究方向。不完备混合型有序信息系统是一种常见的信息系统类型,然而,目前少有增量式属性约简方面的相关研究,针对这一问题,本文在不完备混合型有序信息系统下提出一种对象更新情形的增量式属性约简算法。首先,针对不完备混合型有序信息系统提出了邻域容差优势关系,基于该二元关系建立了一种新的邻域优势粗糙集模型。其次,在其基础上定义了邻域优势条件熵,并利用邻域优势条件熵作为启发式函数设计出一种不完备混合型有序信息系统的非增量式属性约简算法。然后,利用矩阵的形式重构了邻域容差优势关系和邻域优势条件熵,针对不完备混合型有序信息系统对象的动态变化,基于矩阵的计算策略分别研究了邻域优势条件熵随信息系统对象增加和对象减少时的增量式更新。最后,利用邻域优势条件熵的更新机制分别提出了不完备混合型有序信息系统对象增加和对象减少时属性约简的增量式更新算法。实验结果表明:(1)与非增量式算法相比,所提出的增量式算法约简属性数量平均降低了3.6%,分类精度平均提升了2.4%,属性约简的效率平均提升了约10倍;(2)与同类型增量式算法相比,所提出的增量式算法约简属性数量平均降低了9.0%,分类精度平均提升了2.1%,属性约简的平均效率提升了94%。因此,本文所提出增量式算法无论在属性约简结果和属性约简效率上都有着更高的性能。

一种基于新的条件信息熵的高效知识约简算法

分析了在知识约简过程中现有条件信息熵的不足,给出一种新的条件信息熵,由此定义新的属性重要性.将其与基于正区域和基于现有条件信息熵的属性重要性进行比较,结果表明新的属性重要性是一种更准确、更全面的启发信息.以新的属性重要性为启发信息设计约简算法,并给出计算新的条件信息熵的高效算法.理论分析和实验结果表明,与基于现有条件信息熵的约简算法相比,该约简算法时间复杂度较低,且在搜索最小或次优约简方面更优.

基于条件熵的不完备信息系统属性约简算法

在相容关系下定义了三种不完备条件熵——H′条件熵、E′条件熵和I′条件熵,并对它们的性质进行了分析比较,研究发现,H′条件熵和I′条件熵不适用于相容关系下信息观点的约简。利用E′条件熵刻画信息系统中属性的相对重要性,设计了一种新的基于信息论观点的启发式约简算法,它统一了完备信息系统与非完备信息系统中的约简方法。通过实例说明,该算法能得到决策表的相对约简。

一种粗糙集属性约简算法

该文针对RoughSet理论的属性约简进行了研究。利用RoughSet和信息论的相关知识,研究了通过可辨识矩阵求得属性约简集,并利用条件熵来计算属性约简集中属性间的相关性,其平均值最小的属性集即为求得的最佳属性约简的结果。实验证明,它可以取得比较理想的效果。最后利用该文的方法给出了对UCI机器学习数据库的例子的约简结果。

基于信息熵的粗糙集属性约简及其应用

粗集约简对噪声非常敏感,因此寻求噪声干扰下,属性间的准确关系和不确定性关系的表达显得十分重要。将粗糙集理论与信息论理论结合起来,发挥各自优势,取长补短,提出了一种改进的属性约简算法,且在此基础上进行了值约简并应用于超大型船舶的旋回性分析。给出了各因素之间的依赖关系,增比特征,及规则分析。取得了很好效果,对船舶操纵和技术研究有良好的实用价值。

基于信息熵属性约简的航空发动机故障诊断

针对不协调信息条件下的航空发动机故障诊断问题,研究了基于信息熵属性约简的故障诊断方法。首先定义了故障诊断信息系统来描述不协调故障样本数据,针对基本粗糙集模型分类能力不足的问题,引入变精度粗糙集模型处理不协调诊断信息系统;然后针对现有条件熵不能区分不确定性规则的缺陷,提出了变精度条件熵作为属性重要度的度量标准,设计了启发式属性约简算法,提取故障诊断规则。将该方法用于航空发动机故障诊断,验证了该方法可有效处理不协调信息,显著提高了航空发动机故障诊断的准确率。

属性约简准则与约简信息损失的研究

属性约简是粗糙集的重要研究内容,信息熵是度量信息量的方法.在研究绝对约简和几种相对约简的基础上,归纳出属性约简的一般准则.定义了基于条件属性信息熵的属性约简和基于联合熵的属性约简,研究了几种属性约简与绝对约简之间的关系.定义了基于条件属性信息熵的约简信息损失,澄清了属性约简不损失信息的含糊观念,指出了属性约简只是在约简准则意义下不损失信息,在信息熵意义下可能损失信息.为进一步研究粗糙集、粒计算中属性约简与分类夯实了信息论基础.

基于信息熵的一种属性约简算法

文章针对Rough Set理论的核心内容之一属性约简进行了研究。结合信息论的有关知识，研究了在属性约简过程中决策属性集相对条件属性集的条件熵的变化规律，在此基础上提出了新的属性约简算法。实验分析表明，在多数情况下这种算法都能够得到决策表的最小约简，同时还对算法复杂度做了简单的分析。

不完备邻域粗糙集的不确定性度量和属性约简

针对现有的属性约简算法不适合处理数值型属性和符号型属性共同存在的不完备数据,提出了一种拓展不完备邻域粗糙集模型。首先,通过考虑属性值的概率分布来定义缺失属性值之间的距离,可以度量具有混合属性的不完备数据;其次,定义了邻域混合熵来评价属性约简的质量,分析证明了相关的性质定理,并构造了一种基于邻域混合熵的不完备邻域粗糙集属性约简算法;最后从UCI数据集中选取了7组数据进行实验,并分别与基于依赖度的属性约简(ARD)、基于邻域条件熵的属性约简(ARCE)、基于邻域组合测度的属性约简(ARNCM)算法进行了比较。理论分析和实验结果表明,所提算法约简属性比ARD、ARCE、ARNCM分别减少了约1,7,0个,所提算法的分类精度比ARD、ARCE、ARNCM分别提高了约2.5,2.1,0.8个百分点。所提算法不仅能够获得较少的约简属性,同时具有较高的分类精度。

一个有效的基于信息熵的启发式属性约简算法

基于信息熵的属性约简算法都是以信息熵为启发信息设计的,其时间复杂度并不理想.为降低算法的时间复杂度,引入简化决策表的定义,设计了一个求简化决策表的算法,其时间复杂度为O(|C||U|).以快速缩小简化决策表的搜索空间为目的,定义了一个新的、较为合理的、度量属性的信息量,并给出了它的递归计算方法,其时间复杂度为O(|U/C|).同时证明了简化决策表上基于信息量的属性约简与原决策表上基于信息熵的属性约简是等价的.然后以属性的信息量为启发信息,设计了一个基于信息熵的快速属性约简算法,其时间复杂度降为max(O(|C||U|),O(|C|2|U/C|)),并用一个实例说明算法的有效性.实验结果表明新算法不仅具有高效性,且能处理大型决策表.

粗糙集和信息熵的属性约简算法及其应用

阐述粗糙集理论和信息熵的基本概念,并为寻找属性约简的有效方法,提出一种基于粗糙集和信息熵的属性约简算法。在决策表中添加某个属性引起的互信息变化的大小,以反映该属性的重要性,并求相对约简。研究表明,该算法不仅能得到最优的决策规则,而且能够减少信息系统所需的搜索空间,得到更优的属性约简效果。