粗糙集在电力变压器故障诊断中的应用

电力变压器是一种比较复杂的系统,在实际故障诊断中要想获得完备的实验数据比较困难。针对该问题,提出了一种基于粗糙集的电力变压器故障诊断新方法,即分析搜集到的电力变压器历史故障数据,确定条件属性集和决策属性集;对条件属性集进行约简,去除冗余信息,提取关键信息,得到相应的规则集;利用该规则集对电力变压器进行故障诊断。实例分析验证了该方法的正确性和有效性。     

基于粗糙集和信息熵的变压器故障诊断方法

根据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和属性,在分明矩阵中出现的频率作为启发式信息,提出了基于熵和属性频度的属性约简算法.在此基础上把粗糙集自动知识获取的理论应用在电力系统的变压器故障诊断.实例分析表明,该方法有效地减少了故障信息的冗余性,诊断效率高,结果易于理解,在电力系统其它领域可进行类似推广.     

基于分辨矩阵和约简树的增量式属性约简算法

为了对动态变化的决策表进行高效属性约简处理,在改进的分辨矩阵的基础上提出一种基于约简树的增量式属性约简算法IRART,该算法首先根据序贯属性约简算法对原决策表构造约简树,然后求出新增对象的分辨向量,并利用此向量对约简树进行修整,从而快速得到新决策表的所有约简,最后通过示例证明了这种算法的有效性。与传统增量式属性约简算法相比,该算法避免了复杂的逻辑演算,提高了属性约简的更新效率,理论分析表明该算法是有效可行的。     

基于强化正域的属性约简方法

属性约简是粗糙集理论的核心内容之一。通过对多种约简方法进行比较,为了得到更好的结果,在传统基于属性依赖度的约简方法基础上,定义更精确的强化正域概念。通过对边界域的精确划分,得出各条件属性对决策属性的强化依赖度,并用自顶向下的启发式搜索算法得到约简结果。采用UCI标准数据集对基于强化正域约简方法REPR进行测试,约简数据后构建的决策树规模小,分类精度高。实验结果表明,相比于经典方法,REPR能更有效地对决策表进行属性约简。     

一种并行的信息系统属性约简算法*

属性约简是信息系统中的一个重要操作,而分类是属性约简的基础,且直接在大数据集上进行属性约简往往存在效率低下的问题。故以分类为基础提出了一种基于信息熵的信息系统并行属性约简算法。该算法通过信息熵的计算,在属性约简的同时对原信息系统逐层分解成尽量均匀的子表,从而实现了属性约简的并行计算并缩小了搜索空间。对该算法的时间复杂度进行了分析,实验表明,该算法在效率方面优于传统算法。     

基于分形维数的非常规决策表的属性约简

属性约简是粗糙集的一个核心研究课题,但经典属性约简及其延伸算法是基于有决策属性的决策表的属性约简算法,它们对无决策属性的非常规决策表的属性约简无能为力。以粗糙集理论为基础,对无决策属性的非常规决策表从分形维数方面进行研究,提出了一种适用于无决策属性的决策表的启发式属性约简算法。该算法在一定程度上能够解决非常规决策表的属性约简问题,进一步扩展了粗糙集理论的应用范围。实例表明该算法是有效可行的。     

基于聚类率的决策表属性约简

根据经典粗糙集方法,在对可约简决策表进行属性约简时可能出现核不存在无法找到起点属性从而无法约简的问题。针对该问题,提出了基于聚类率的属性约简方法。计算决策表的区分矩阵,以属性区分度为基础,在属性区分度相同的情况下,利用聚类率修正属性重要度,保证起点属性存在的必然性,从而完成起点属性的求取并获得决策表的属性约简。实验分析表明,方法可以保证可约简决策表中起点属性的计算,且利用该方法获得的约简结果保持了较高的决策准确率,是有效可行的。     

基于粒计算与区分能力的属性约简算法?

传统的属性约简方法将整个数据集一次性装入内存,很难适应大数据背景下的数据分析。为此文中提出基于粒计算与区分能力的属性约简算法。该算法运用统计学中的分层抽样技术,拆分原始大数据集为多个样本子集(粒),在每个粒上运用属性的区分能力进行属性约简,最后将各粒约简结果进行加权融合,得到原始大数据集的属性约简结果。实验表明该算法对海量数据集进行属性约简的可行性和高效性。     

基于邻域信息熵度量数值属性快速约简算法

阐述邻域粗糙集和邻域信息熵的基本定义及性质,为避免数值属性信息系统属性约简过程中,属性离散化造成特征信息的丢失,提出一种新的基于邻域信息熵度量数值属性约简算法。扩展邻域信息系统核属性集生成约简属性集,邻域信息熵度量不仅关注约简属性集正域变化,而且考察负域样本空间约简属性邻域等价类在决策属性划分的分布,具备更好的邻域关系度量细粒度。实验表明,对比邻域粗糙集近似度量、邻域有效信息率度量、邻域软间隔度量的属性约简方法,该算法能有效进行邻域信息系统属性约简的同时,也保持了约简属性集更好的分类精度。     

基于决策表的区分矩阵增量属性约简算法

对于决策表中存在对象动态变化的现象,当利用静态的属性约简算法处理这类决策表时算法效率并不理想,为了有效提高增量属性约简算法的效率,对决策表进行了简化,并证明了基于简化区分矩阵的属性约简与基于区分矩阵的属性约简是一致的,在利用原的属性约简的基础上,提出了一种基于决策表的区分矩阵增量属性约简算法,通过实例分析说明算法的有效性和可行性。     

A study of the relation of mobile device attributes with the user-perceived quality of Android apps

The number of mobile applications (apps) and mobile devices has increased considerably over the past few years. Online app markets, such as the Google Play Store, use a star-rating mechanism to quantify the user-perceived quality of mobile apps. Users may rate apps on a five point (star) scale where a five star-rating is the highest rating. Having considered the importance of a high star-rating to the success of an app, recent studies continue to explore the relationship between the app attributes, such as User Interface (UI) complexity, and the user-perceived quality. However, the user-perceived quality reflects the users’ experience using an app on a particular mobile device. Hence, the user-perceived quality of an app is not solely determined by app attributes. In this paper, we study the relation of both device attributes and app attributes with the user-perceived quality of Android apps from the Google Play Store. We study 20 device attributes, such as the CPU and the display size, and 13 app attributes, such as code size and UI complexity. Our study is based on data from 30 types of Android mobile devices and 280 Android apps. We use linear mixed effect models to identify the device attributes and app attributes with the strongest relationship with the user-perceived quality. We find that the code size has the strongest relationship with the user-perceived quality. However, some device attributes, such as the CPU, have stronger relationships with the user-perceived quality than some app attributes, such as the number of UI inputs and outputs of an app. Our work helps both device manufacturers and app developers. Manufacturers can focus on the attributes that have significant relationships with the user-perceived quality. Moreover, app developers should be careful about the devices for which they make their apps available because the device attributes have a strong relationship with the ratings that users give to apps.     

ISAD:一种新的基于属性距离和的孤立点检测算法

孤立点是数据对象在某些属性(维)上波动形成的.由此,本文提出了关键属性的概念,用于描述影响数据稳定性的属性.在真实数据集中,只有一部分属性是能够决定某数据是否是孤立点的关键属性.由此,本文提出了关键属性隶属度的定义及其求解算法,并在此基础上提出了一种新的基于属性距离和的孤立点检测算法.实验结果表明,该算法较基于单元的算法在效率及雏数可扩展方面均有显著提高.     

一种基于属性重要性的启发式约简算法

属性约简是知识发现中的关键问题之一.为了能够有效地获取决策表中条件属性集的最小相对约简,本文首先利用代数方法描述决策表中的属性的重要性,提出了限制正域的概念,得到了关于限制正域的若干结果,并据此提出一种改进的属性约简算法,即以属性核为起点并结合算子,通过向属性核不断添加重要程度最大的属性,并利用已求得的正区域和限制正域使处理数据的范围不断缩小从而减少求约简的时间. 该算法能够节省得到决策表的最小约简的时间并能得到所有相对约简.实例分析也验证了该算法的有效性.     

Revealing the dual importance and Kano type of attributes through customer review analytics

Identifying which attributes of a product are important to customers and clarifying how the attributes affect customer satisfaction are critical for a firm to survive and succeed in the market. To assist in characterizing the impacts of various attributes and prioritizing the attributes for design and marketing purposes, this paper proposes a novel review-analytics framework, called importance-Kano (I-Kano) analysis. I-Kano analysis holistically assesses the impacts of various attributes from three different perspectives that potentially may conflict with each other, i.e., appearance (stated importance), significance (derived importance), and Kano type. By fusing term-frequency and sentiment analyses of online reviews with conjoint analysis, the I-Kano analysis simultaneously identifies the dual importance (appearance and significance) and Kano type of an attribute. As the final deliverable of the I-Kano analysis, a new visualization scheme, called the I-Kano matrix, is proposed, which is the first attempt to integrate the dual importance and Kano type of multiple attributes in a single chart. The I-Kano matrix facilitates an intuitive interpretation of the multidimensional impacts of various attributes and supports the aggregation and comparison of the results from different market segments. Through the I-Kano analysis, the attributes of great importance in a market segment, which are useful for developing products and planning marketing promotions, can be identified. In addition, the I-Kano analysis can identify the segments of the market in which a certain attribute has greater relative importance, which is helpful in the design differentiation, targeting, and differentiated marketing of products. To demonstrate and validate the I-Kano analysis, an illustrative case study is described with an example of online hotel reviews.     

Assessing the importance of autistic attributes for autism screening

Autistic Spectrum Disorder (ASD) is a cognitive disease which leads to the loss of linguistic, communicative, cognitive, and social skills and abilities. Patients with ASD have diverse troubles such as sleeping problems. The role of genetic and environmental factors is of great importance in its pathophysiology. Early diagnosis provides an improved overall mental health of the patients. This study aimed to identify the important attributes for the best detection of this disorder in children, adolescents and adults. To achieve this aim, Recursive Feature Elimination and Stability Selection methods that consider important attributes for target class were proposed. The attributes collected from autism screening methods and other attributes such as age and gender were examined for the disease. The results verified the combining of feature selection method and machine learning algorithm within the frame of accuracy, sensitivity and specificity evaluation metrics.     

变精度粗糙集的属性核和最小属性约简算法

文中深入研究了变精度粗糙集的属性约简问题,给出了3种属性约简的概念,针对不同概念的属性约简,分别提出了两种不同的求解变精度粗糙集最小属性约简算法:基于容差矩阵和属性核的最小约简.提出了变精度粗糙集的属性核思想,对其进行了形式化描述,说明了变精度粗糙集的属性核真正具备了核的本质特征,从而更深层地提出了基于属性核的启发式约简以求解最小约简.理论分析和实例表明,所提出的两种最小约简算法可以减小属性约简的搜索空间,提高约简的效率,使得变精度粗糙集的属性约简具有了实用性.     

基于BP神经网络的属性匹配方法研究

为了实现异构数据库的数据共享，关键的问题就是要找出数据库间的相同属性。目前主要采用的方法是通过比较所有的属性来实现属性的相似性匹配，但是当同一属性用不同数据类型表示时，由于描述属性的元数据信息和取值信息的极大差异性，这些方法就不能找出相同的属性。并且将不同数据类型描述的属性放在一起匹配，还会造成属性数据之间的干扰，影响匹配结果的准确性。为此，本文提出一种基于BP神经网络的二步检查法属性匹配算法。该算法中属性首先根据数据类型进行分类，然后用分类后的属性集分别多次训练神经网络，并对每次的匹配结果求交集作为最终的属性匹配结果，进行两阶段检查，即二步检查法。该算法能有效地消除不一致信息的干扰，降低神经网络的规模，并且可以实现不同数据类型的属性集之间属性匹配过程的并行计算。实验结果显示本文提出的方法能明显地提高系统的运行效率、属性匹配的查准率和查全率。     

An interval-valued 2-tuple linguistic group decision-making model based on the Choquet integral operator

The Choquet integral (IL) operator is an effective approach for handling interdependence among decision attributes in complex decision-making problems. However, the fuzzy measures of attributes and attribute sets required by IL are difficult to achieve directly, which limits the application of IL. This paper proposes a new method for determining fuzzy measures of attributes by extending Marichal's concept of entropy for fuzzy measure. To well represent the assessment information, interval-valued 2-tuple linguistic context is utilised to represent information. Then, we propose a Choquet integral operator in an interval-valued 2-tuple linguistic environment, which can effectively handle the correlation between attributes. In addition, we apply these methods to solve multi-attribute group decision-making problems. The feasibility and validity of the proposed operator is demonstrated by comparisons with other models in illustrative example part.