Blind source separation by weighted K-means clustering

Blind separation of sparse sources （BSSS） is discussed. The BSSS method based on the conventional K-means clustering is very fast and is also easy to implement. However, the accuracy of this method is generally not satisfactory. The contribution of the vector x（t） with different modules is theoretically proved to be unequal, and a weighted K-means clustering method is proposed on this grounds. The proposed algorithm is not only as fast as the conventional K-means clustering method, but can also achieve considerably accurate results, which is demonstrated by numerical experiments.

Fault Diagnosis Model Based on Fuzzy Support Vector Machine Combined with Weighted Fuzzy Clustering

A fault diagnosis model is proposed based on fuzzy support vector machine (FSVM) combined with fuzzy clustering (FC).Considering the relationship between the sample point and non-self class,FC algorithm is applied to generate fuzzy memberships.In the algorithm,sample weights based on a distribution density function of data point and genetic algorithm (GA) are introduced to enhance the performance of FC.Then a multi-class FSVM with radial basis function kernel is established according to directed acyclic graph algorithm,the penalty factor and kernel parameter of which are optimized by GA.Finally,the model is executed for multi-class fault diagnosis of rolling element bearings.The results show that the presented model achieves high performances both in identifying fault types and fault degrees.The performance comparisons of the presented model with SVM and distance-based FSVM for noisy case demonstrate the capacity of dealing with noise and generalization.

Extreme scenario extraction of a grid with large scale wind power integration by combined entropy-weighted clustering method

Large-scale integration of wind power into a power system introduces uncertainties to its operation and planning,making the power system operation scenario highly diversified and variable.In conventional power system planning,some key operation modes and most critical scenarios are typically analyzed to identify the weak and high-risk points in grid operation.While these scenarios may not follow traditional empirical patterns due to the introduction of large-scale wind power.In this paper,we propose a weighted clustering method to quickly identify a system’s extreme operation scenarios by considering the temporal variations and correlations between wind power and load to evaluate the stability and security for system planning.Specifically,based on an annual time-series data of wind power and load,a combined weighted clustering method is used to pick the typical scenarios of power grid operation,and the edge operation points far from the clustering center are extracted as the extreme scenarios.The contribution of fluctuations and capacities of different wind farms and loads to extreme scenarios are considered in the clustering process,to further improve the efficiency and rationality of the extreme-scenario extraction.A set of case studies was used to verify the performance of the method,providing an intuitive understanding of the extreme scenario variety under wind power integration.

NEW SHADOWED C-MEANS CLUSTERING WITH FEATURE WEIGHTS

Partition-based clustering with weighted feature is developed in the framework of shadowed sets. The objects in the core and boundary regions, generated by shadowed sets-based clustering, have different impact on the prototype of each cluster. By integrating feature weights, a formula for weight calculation is introduced to the clustering algorithm. The selection of weight exponent is crucial for good result and the weights are updated iteratively with each partition of clusters. The convergence of the weighted algorithms is given, and the feasible cluster validity indices of data mining application are utilized. Experimental results on both synthetic and real-life numerical data with different feature weights demonstrate that the weighted algorithm is better than the other unweighted algorithms.

基于Weighted-slope One的用户聚类推荐算法研究

针对传统协同过滤推荐算法存在的数据稀疏性以及实时性差的问题,提出一种基于Weighted-slope One的用户聚类推荐算法。该算法首先利用Weighted-slope One算法的思想对初始的用户-评分矩阵进行有效填充,降低数据的稀疏性;然后,结合初始聚类中心优化改进的K-means方法对用户进行聚类,生成相似用户集合,以缩小目标用户搜索最近邻的范围;最后,结合目标用户所属的聚类,利用基于用户的协同过滤算法搜索最近邻居,为目标用户推荐对应的产品。仿真实验结果表明,改进算法可以显著降低数据的稀疏度,同时提升推荐的准确性和实时性。

Construction mechanism of whitenization weight function and its application in grey clustering evaluation

The clustering evaluation can be used to scientifically classify the objects to be evaluated according to the information aggregation of various evaluation rules. In grey weighted clustering evaluation, the index clustering rule relies on the construction of the whitenization weight function, while the existing construction method of the linear function lacks the construction mechanism analysis and validity explanation. A normative construction principle is put forward by analyzing the construction mechanism of the function. Through proving the normative principle of the function,the basic modal function(BMF) is proposed and characterized by different function forms. Then, a new type of the whitenization weight function and its grey clustering evaluation model algorithm are given by studying the mechanism and nature of the construction of different forms of the function. Finally, the comparative study for self-innovation capability of defense science and technology industry(DSTI) is taken as an example. The results show that the different construction ways of the function have an effect on the clustering result. The proposed construction mechanism can better explain the index clustering rules and evaluation effectiveness,which will perfect the theoretical system of grey clustering evaluation and be applied to practice effectively.

Weighted Markov chains for forecasting and analysis in Incidence of infectious diseases in jiangsu Province,China

This paper first applies the sequential cluster method to set up the classification standard of infectious disease incidence state based on the fact that there are many uncertainty characteristics in the incidence course.Then the paper presents a weighted Markov chain,a method which is used to predict the future incidence state.This method assumes the standardized self-coefficients as weights based on the special characteristics of infectious disease incidence being a dependent stochastic variable.It also analyzes the characteristics of infectious diseases incidence via the Markov chain Monte Carlo method to make the long-term benefit of decision optimal.Our method is successfully validated using existing incidents data of infectious diseases in Jiangsu Province.In summation,this paper proposes ways to improve the accuracy of the weighted Markov chain,specifically in the field of infection epidemiology.

Mining Social Groups with Weighted Similarity in Campus Wireless Network

With the popularity of wireless networks and the prevalence of personal mobile computing devices, understanding the characteristic of wireless network users is of great significance to the network performance. In this study, system logs from two universities, Dartmouth College and Shanghai Jiao Tong University(SJTU), were mined and analyzed. Every user's log was represented by a user profile. A novel weighted social similarity was proposed to quantify the resemblance of users considering influence of location visits. Based on the similarity, an unsupervised learning method was applied to cluster users. Though environment parameters are different, two universities both form many social groups with Pareto distribution of similarity and exponential distribution of group sizes. These findings are very important to the research of wireless network and social network .

Optimization Study of Outburst Prevention Measures for Tuzhu Coal Mine Based on Fixed Weight Clustering Analysis

Affected by many involved factors, different dimensions, data with large difference, incomplete information and so on, the most optimal selection of regional outburst prevention measures for outburst mine has become a complicated system project. The traditional way of outburst prevention measure selection belongs to qualitative method, which may cause high-cost of gas control, huge quantities of drilling work, long construction time and even secondary disaster. To solve the above-mentioned problems, in light of occurrence status of coal seam gas in No. 21 mining area of Jinzhushan Tuzhu Mine, through grey fixed weight clustering theory and a combination method of qualitative and quantitative analysis, the judging model with multi-objective classification for optimization of outburst prevention measures was established. The three weight coefficients of outburst prevention technology scheme are sorted, in order to determine the advantages and disadvantages of each outburst prevention technology scheme under the comprehensive evaluation of multi-target. Finally, the problem of quantitative selection for regional outburst prevention technology scheme is solved under the situation of multi-factor mode and incomplete information, which provides reasonable and effective technical measures for prevention of coal and gas outburst disaster.

Deep Learning and Tensor-Based Multiple Clustering Approaches for Cyber-Physical-Social Applications

The study delves into the expanding role of network platforms in our daily lives, encompassing various mediums like blogs, forums, online chats, and prominent social media platforms such as Facebook, Twitter, and Instagram. While these platforms offer avenues for self-expression and community support, they concurrently harbor negative impacts, fostering antisocial behaviors like phishing, impersonation, hate speech, cyberbullying, cyberstalking, cyberterrorism, fake news propagation, spamming, and fraud. Notably, individuals also leverage these platforms to connect with authorities and seek aid during disasters. The overarching objective of this research is to address the dual nature of network platforms by proposing innovative methodologies aimed at enhancing their positive aspects and mitigating their negative repercussions. To achieve this, the study introduces a weight learning method grounded in multi-linear attribute ranking. This approach serves to evaluate the significance of attribute combinations across all feature spaces. Additionally, a novel clustering method based on tensors is proposed to elevate the quality of clustering while effectively distinguishing selected features. The methodology incorporates a weighted average similarity matrix and optionally integrates weighted Euclidean distance, contributing to a more nuanced understanding of attribute importance. The analysis of the proposed methods yields significant findings. The weight learning method proves instrumental in discerning the importance of attribute combinations, shedding light on key aspects within feature spaces. Simultaneously, the clustering method based on tensors exhibits improved efficacy in enhancing clustering quality and feature distinction. This not only advances our understanding of attribute importance but also paves the way for more nuanced data analysis methodologies. In conclusion, this research underscores the pivotal role of network platforms in contemporary society, emphasizing their potential for both positive contributions and adverse consequences. The proposed methodologies offer novel approaches to address these dualities, providing a foundation for future research and practical applications. Ultimately, this study contributes to the ongoing discourse on optimizing the utility of network platforms while minimizing their negative impacts.

A Mixed Mechanism of Weighted-Driven and Inner Selection in Networks

For most networks, the weight of connection is changing with their attachment and inner affinity. By introducing a mixed mechanism of weighted-driven and inner selection, the model exhibits wide range power-law distributions of node strength and edge weight, and the exponent can be adjusted by not only the parameter δ but also the probability q. Furthermore, we investigate the weighted average shortest distance, clustering coefficient, and the correlation of our network. In addition, the weighted assortativity coefficient which characterizes important information of weighted topological networks has been discussed, but the variation of coefficients is much smaller than the former researches.

Unsupervised Functional Data Clustering Based on Adaptive Weights

In recent years, functional data has been widely used in finance, medicine, biology and other fields. The current clustering analysis can solve the problems in finite-dimensional space, but it is difficult to be directly used for the clustering of functional data. In this paper, we propose a new unsupervised clustering algorithm based on adaptive weights. In the absence of initialization parameter, we use entropy-type penalty terms and fuzzy partition matrix to find the optimal number of clusters. At the same time, we introduce a measure based on adaptive weights to reflect the difference in information content between different clustering metrics. Simulation experiments show that the proposed algorithm has higher purity than some algorithms.

A comprehensive evaluation of enterprise emergency management capacity （EEMC） based on variable weight gray cluster

On the basis of the initial definition of Enterprise Emergency Management Capacity（EEMC）, the paper has established evaluation index system of EEMC, and provided a method to calculate index weight, with the regard to subjectivity existing in the comprehensive evaluation of EEMC multi-indicators, in accordance with the principle of Variable weight Gray Cluster, which makes the weight of indicators generate automatically in the evaluation process and not judged by human, thus decreasing subjective factors during the evaluation.

基于WIM的地质灾害易发性评价--以遵化市为例

以遵化市为例,运用聚类分析方法确定坡高、坡度、坡向、工程地质岩组、地质构造、归一化植被指数(NDVI)、水系、道路8项因子组成地质灾害易发性评价指标体系,运用层析分析法(AHP)与信息量(IM)模型耦合后加权信息量(WIM)模型对研究区开展地质灾害易发性评价。结果表明:研究区高易发区、中易发区、低易发区和非易发区面积(占比)分别为60.94 km^(2)(4.007%)、487.87 km^(2)(32.076%)、330.93 km^(2)(21.757%)、641.26 km^(2)(42.160%);经受试者工作曲线验证,WIM评价结果AUC值为0.853,高于IM评价结果(AUC=0.712),表明WIM模型评价结果精度更高。研究可为本区域地质灾害评价、防灾减灾工作提供参考依据。

运用模糊隶属度进行土壤属性制图的研究--以黑龙江鹤山农场研究区为例

通过传统土壤类型图所得的土壤属性图已不能满足精准农业和生态环境模型所需土壤属性的精度,而目前应用较多的统计方法和地统计方法均存在一定的局限性。鉴于此,本文探索了一种采用模糊聚类获取模糊隶属度进行土壤属性制图的方法。首先,采用模糊c均值聚类(Fuzzyc-means clustering,FCM)方法对环境因子进行聚类,通过野外采样(称为建模点)建立土壤-环境关系知识;然后,计算区域内各像元点对土壤类型的模糊隶属度;最后,对模糊隶属度采用加权平均的方法获取土壤属性值。将该方法应用于黑龙江鹤山农场老莱河流域的研究小区,以土体厚度和表层有机质为例进行土壤属性制图。为了评价该方法的有效性,将其与采用环境因子所建立的多元线性回归模型进行比较,通过野外验证点集评价两种模型所得的土壤属性,评价指标为观测值和预测值的相关系数、平均绝对误差(MAE)、均方根误差(RMSE)和准确度(AC)。结果表明,尽管通过建模点建立的多元线性回归方程R2较大,但该方程并不适用于研究区内的其他样本点,这表明多元线性回归方法在该区具有一定的局限性。与之相比,模糊隶属度加权平均的方法则可以通过较少的建模点得到更好的预测效果。

Improved method for the feature extraction of laser scanner using genetic clustering

Feature extraction of range images provided by ranging sensor is a key issue of pattern recognition. To automatically extract the environmental feature sensed by a 2D ranging sensor laser scanner, an improved method based on genetic clustering VGA-clustering is presented. By integrating the spatial neighbouring information of range data into fuzzy clustering algorithm, a weighted fuzzy clustering algorithm （WFCA） instead of standard clustering algorithm is introduced to realize feature extraction of laser scanner. Aimed at the unknown clustering number in advance, several validation index functions are used to estimate the validity of different clustering algorithms and one validation index is selected as the fitness function of genetic algorithm so as to determine the accurate clustering number automatically. At the same time, an improved genetic algorithm IVGA on the basis of VGA is proposed to solve the local optimum of clustering algorithm, which is implemented by increasing the population diversity and improving the genetic operators of elitist rule to enhance the local search capacity and to quicken the convergence speed. By the comparison with other algorithms, the effectiveness of the algorithm introduced is demonstrated.

Weighting indicators of building energy efficiency assessment taking account of experts' priority

Analytic hierarchy process(Group AHP) is combined with two different methods of assigning experts' priority to weight indicators in building energy efficiency assessment.One is to assign the experts' priority averagely,and the other is to use cluster analysis to assign experts' priority.The results show that,1) Different expert's priority assigns result in great different weights of indicators in building energy efficiency assessment,therefore,the method of assigning experts' priority should be taken into account carefully while weighting indicators of building energy efficiency assessment using Group AHP;2) Three indicators are found to be overwhelmingly important in residential building energy efficiency assessment in the hot summer and cold winter zone in China.They are 'Outdoor & indoor shadow','Heating & air-conditioning facilities' and 'Insulation of envelope';3) The method combining cluster analysis with Group AHP to weight indicator of building energy efficiency assessment has the advantage of finding overwhelming important indicator,whereas,some less important indicators have a tendency to be ignored.A useful reference is provided for building energy conservation including policy revision and energy efficient residential building design.

Assessing the Eutrophication of Shengzhong Reservoir Based on Grey Clustering Method

Reservoir water environment is a grey system. The grey clustering method is applied to assessing the reservoir water enviromnent to establish a relatively complete model suitable for the reservoir eutrophication evaluation and appropriately evaluate the quality of reservoir water, providing evidence for reservoir management. According to China＇s lakes and reservoir eutrophication criteria and the characteristics of China＇s eutrophication, as well as certain evaluation indices, the degree of eutrophication is classified into six categories with the utilization of grey classified whitening weight function to represent the boundaries of classification, to determine the clustering weight and clustering coefficient of each index in grey classifications, and the classification of each clustering object. The comprehensive evaluation of reservoir eutrophication is established on such a foundation, with Sichuan Shengzhong Reservoir as the survey object and the analysis of the data attained by several typical monitoring points there in 2006. It is found that eutrophication of Tiebian Power Generation Station, Guoyuanchang and Dashiqiao Bridge is the heaviest, Tielusi and Qinggangya the second, and Lijiaba the least. The eutrophication of this reservoir is closely relevant to the irrational exploitation in its surrounding areas, especially to the aggravation of the non-point source pollution and the increase of net-culture fishing. Therefore, it is feasible to use grey clustering in environment quality evaluation, and the point lies in the correct division of grey whitening function

Energy efficient clustering algorithm based on neighbors for wireless sensor networks

In this paper, an energy efficient clustering algorithm based on neighbors （EECABN） for wireless sensor networks is proposed. In the algorithm, an optimized weight of nodes is introduced to determine the priority of clustering procedure. As improvement, the weight is a measurement of energy and degree as usual, and even associates with distance from neighbors, distance to the sink node, and other factors. To prevent the low energy nodes being exhausted with energy, the strong nodes should have more opportunities to act as cluster heads during the clustering procedure. The simulation results show that the algorithm can effectively prolong whole the network lifetime. Especially at the early stage that some nodes in the network begin to die, the process can be postponed by using the algorithm.