Cluster analysis is one of the major data analysis methods widely used for many practical applications in emerging areas of data mining. A good clustering method will produce high quality clusters with high intra-clus...Cluster analysis is one of the major data analysis methods widely used for many practical applications in emerging areas of data mining. A good clustering method will produce high quality clusters with high intra-cluster similarity and low inter-cluster similarity. Clustering techniques are applied in different domains to predict future trends of available data and its uses for the real world. This research work is carried out to find the performance of two of the most delegated, partition based clustering algorithms namely k-Means and k-Medoids. A state of art analysis of these two algorithms is implemented and performance is analyzed based on their clustering result quality by means of its execution time and other components. Telecommunication data is the source data for this analysis. The connection oriented broadband data is given as input to find the clustering quality of the algorithms. Distance between the server locations and their connection is considered for clustering. Execution time for each algorithm is analyzed and the results are compared with one another. Results found in comparison study are satisfactory for the chosen application.展开更多
Data stream clustering is integral to contemporary big data applications.However,addressing the ongoing influx of data streams efficiently and accurately remains a primary challenge in current research.This paper aims...Data stream clustering is integral to contemporary big data applications.However,addressing the ongoing influx of data streams efficiently and accurately remains a primary challenge in current research.This paper aims to elevate the efficiency and precision of data stream clustering,leveraging the TEDA(Typicality and Eccentricity Data Analysis)algorithm as a foundation,we introduce improvements by integrating a nearest neighbor search algorithm to enhance both the efficiency and accuracy of the algorithm.The original TEDA algorithm,grounded in the concept of“Typicality and Eccentricity Data Analytics”,represents an evolving and recursive method that requires no prior knowledge.While the algorithm autonomously creates and merges clusters as new data arrives,its efficiency is significantly hindered by the need to traverse all existing clusters upon the arrival of further data.This work presents the NS-TEDA(Neighbor Search Based Typicality and Eccentricity Data Analysis)algorithm by incorporating a KD-Tree(K-Dimensional Tree)algorithm integrated with the Scapegoat Tree.Upon arrival,this ensures that new data points interact solely with clusters in very close proximity.This significantly enhances algorithm efficiency while preventing a single data point from joining too many clusters and mitigating the merging of clusters with high overlap to some extent.We apply the NS-TEDA algorithm to several well-known datasets,comparing its performance with other data stream clustering algorithms and the original TEDA algorithm.The results demonstrate that the proposed algorithm achieves higher accuracy,and its runtime exhibits almost linear dependence on the volume of data,making it more suitable for large-scale data stream analysis research.展开更多
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 Inst...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.展开更多
Implementing machine learning algorithms in the non-conducive environment of the vehicular network requires some adaptations due to the high computational complexity of these algorithms.K-clustering algorithms are sim...Implementing machine learning algorithms in the non-conducive environment of the vehicular network requires some adaptations due to the high computational complexity of these algorithms.K-clustering algorithms are simplistic,with fast performance and relative accuracy.However,their implementation depends on the initial selection of clusters number(K),the initial clusters’centers,and the clustering metric.This paper investigated using Scott’s histogram formula to estimate the K number and the Link Expiration Time(LET)as a clustering metric.Realistic traffic flows were considered for three maps,namely Highway,Traffic Light junction,and Roundabout junction,to study the effect of road layout on estimating the K number.A fast version of the PAM algorithm was used for clustering with a modification to reduce time complexity.The Affinity propagation algorithm sets the baseline for the estimated K number,and the Medoid Silhouette method is used to quantify the clustering.OMNET++,Veins,and SUMO were used to simulate the traffic,while the related algorithms were implemented in Python.The Scott’s formula estimation of the K number only matched the baseline when the road layout was simple.Moreover,the clustering algorithm required one iteration on average to converge when used with LET.展开更多
Although many multi-view clustering(MVC) algorithms with acceptable performances have been presented, to the best of our knowledge, nearly all of them need to be fed with the correct number of clusters. In addition, t...Although many multi-view clustering(MVC) algorithms with acceptable performances have been presented, to the best of our knowledge, nearly all of them need to be fed with the correct number of clusters. In addition, these existing algorithms create only the hard and fuzzy partitions for multi-view objects,which are often located in highly-overlapping areas of multi-view feature space. The adoption of hard and fuzzy partition ignores the ambiguity and uncertainty in the assignment of objects, likely leading to performance degradation. To address these issues, we propose a novel sparse reconstructive multi-view evidential clustering algorithm(SRMVEC). Based on a sparse reconstructive procedure, SRMVEC learns a shared affinity matrix across views, and maps multi-view objects to a 2-dimensional humanreadable chart by calculating 2 newly defined mathematical metrics for each object. From this chart, users can detect the number of clusters and select several objects existing in the dataset as cluster centers. Then, SRMVEC derives a credal partition under the framework of evidence theory, improving the fault tolerance of clustering. Ablation studies show the benefits of adopting the sparse reconstructive procedure and evidence theory. Besides,SRMVEC delivers effectiveness on benchmark datasets by outperforming some state-of-the-art methods.展开更多
Clustering a social network is a process of grouping social actors into clusters where intra-cluster similarities among actors are higher than inter-cluster similarities. Clustering approaches, i.e. , k-medoids or hie...Clustering a social network is a process of grouping social actors into clusters where intra-cluster similarities among actors are higher than inter-cluster similarities. Clustering approaches, i.e. , k-medoids or hierarchical, use the distance function to measure the dissimilarities among actors. These distance functions need to fulfill various properties, including the triangle inequality (TI). However, in some cases, the triangle inequality might be violated, impacting the quality of the resulting clusters. With experiments, this paper explains how TI violates while performing traditional clustering techniques: k-medoids, hierarchical, DENGRAPH, and spectral clustering on social networks and how the violation of TI affects the quality of the resulting clusters.展开更多
Stochastic unit commitment is one of the most powerful methods to address uncertainty. However, the existingscenario clustering technique for stochastic unit commitment cannot accurately select representative scenario...Stochastic unit commitment is one of the most powerful methods to address uncertainty. However, the existingscenario clustering technique for stochastic unit commitment cannot accurately select representative scenarios,which threatens the robustness of stochastic unit commitment and hinders its application. This paper providesa stochastic unit commitment with dynamic scenario clustering based on multi-parametric programming andBenders decomposition. The stochastic unit commitment is solved via the Benders decomposition, which decouplesthe primal problem into the master problem and two types of subproblems. In the master problem, the committedgenerator is determined, while the feasibility and optimality of generator output are checked in these twosubproblems. Scenarios are dynamically clustered during the subproblem solution process through the multiparametric programming with respect to the solution of the master problem. In other words, multiple scenariosare clustered into several representative scenarios after the subproblem is solved, and the Benders cut obtainedby the representative scenario is generated for the master problem. Different from the conventional stochasticunit commitment, the proposed approach integrates scenario clustering into the Benders decomposition solutionprocess. Such a clustering approach could accurately cluster representative scenarios that have impacts on theunit commitment. The proposed method is tested on a 6-bus system and the modified IEEE 118-bus system.Numerical results illustrate the effectiveness of the proposed method in clustering scenarios. Compared withthe conventional clustering method, the proposed method can accurately select representative scenarios whilemitigating computational burden, thus guaranteeing the robustness of unit commitment.展开更多
Multi-view Subspace Clustering (MVSC) emerges as an advanced clustering method, designed to integrate diverse views to uncover a common subspace, enhancing the accuracy and robustness of clustering results. The signif...Multi-view Subspace Clustering (MVSC) emerges as an advanced clustering method, designed to integrate diverse views to uncover a common subspace, enhancing the accuracy and robustness of clustering results. The significance of low-rank prior in MVSC is emphasized, highlighting its role in capturing the global data structure across views for improved performance. However, it faces challenges with outlier sensitivity due to its reliance on the Frobenius norm for error measurement. Addressing this, our paper proposes a Low-Rank Multi-view Subspace Clustering Based on Sparse Regularization (LMVSC- Sparse) approach. Sparse regularization helps in selecting the most relevant features or views for clustering while ignoring irrelevant or noisy ones. This leads to a more efficient and effective representation of the data, improving the clustering accuracy and robustness, especially in the presence of outliers or noisy data. By incorporating sparse regularization, LMVSC-Sparse can effectively handle outlier sensitivity, which is a common challenge in traditional MVSC methods relying solely on low-rank priors. Then Alternating Direction Method of Multipliers (ADMM) algorithm is employed to solve the proposed optimization problems. Our comprehensive experiments demonstrate the efficiency and effectiveness of LMVSC-Sparse, offering a robust alternative to traditional MVSC methods.展开更多
To solve the problems of a few optical fibre line fault samples and the inefficiency of manual communication optical fibre fault diagnosis,this paper proposes a communication optical fibre fault diagnosis model based ...To solve the problems of a few optical fibre line fault samples and the inefficiency of manual communication optical fibre fault diagnosis,this paper proposes a communication optical fibre fault diagnosis model based on variational modal decomposition(VMD),fuzzy entropy(FE)and fuzzy clustering(FC).Firstly,based on the OTDR curve data collected in the field,VMD is used to extract the different modal components(IMF)of the original signal and calculate the fuzzy entropy(FE)values of different components to characterize the subtle differences between them.The fuzzy entropy of each curve is used as the feature vector,which in turn constructs the communication optical fibre feature vector matrix,and the fuzzy clustering algorithm is used to achieve fault diagnosis of faulty optical fibre.The VMD-FE combination can extract subtle differences in features,and the fuzzy clustering algorithm does not require sample training.The experimental results show that the model in this paper has high accuracy and is relevant to the maintenance of communication optical fibre when compared with existing feature extraction models and traditional machine learning models.展开更多
The k-means algorithm is a popular data clustering technique due to its speed and simplicity. However, it is susceptible to issues such as sensitivity to the chosen seeds, and inaccurate clusters due to poor initial s...The k-means algorithm is a popular data clustering technique due to its speed and simplicity. However, it is susceptible to issues such as sensitivity to the chosen seeds, and inaccurate clusters due to poor initial seeds, particularly in complex datasets or datasets with non-spherical clusters. In this paper, a Comprehensive K-Means Clustering algorithm is presented, in which multiple trials of k-means are performed on a given dataset. The clustering results from each trial are transformed into a five-dimensional data point, containing the scope values of the x and y coordinates of the clusters along with the number of points within that cluster. A graph is then generated displaying the configuration of these points using Principal Component Analysis (PCA), from which we can observe and determine the common clustering patterns in the dataset. The robustness and strength of these patterns are then examined by observing the variance of the results of each trial, wherein a different subset of the data keeping a certain percentage of original data points is clustered. By aggregating information from multiple trials, we can distinguish clusters that consistently emerge across different runs from those that are more sensitive or unlikely, hence deriving more reliable conclusions about the underlying structure of complex datasets. Our experiments show that our algorithm is able to find the most common associations between different dimensions of data over multiple trials, often more accurately than other algorithms, as well as measure stability of these clusters, an ability that other k-means algorithms lack.展开更多
Wireless sensor networks(WSN)gather information and sense information samples in a certain region and communicate these readings to a base station(BS).Energy efficiency is considered a major design issue in the WSNs,a...Wireless sensor networks(WSN)gather information and sense information samples in a certain region and communicate these readings to a base station(BS).Energy efficiency is considered a major design issue in the WSNs,and can be addressed using clustering and routing techniques.Information is sent from the source to the BS via routing procedures.However,these routing protocols must ensure that packets are delivered securely,guaranteeing that neither adversaries nor unauthentic individuals have access to the sent information.Secure data transfer is intended to protect the data from illegal access,damage,or disruption.Thus,in the proposed model,secure data transmission is developed in an energy-effective manner.A low-energy adaptive clustering hierarchy(LEACH)is developed to efficiently transfer the data.For the intrusion detection systems(IDS),Fuzzy logic and artificial neural networks(ANNs)are proposed.Initially,the nodes were randomly placed in the network and initialized to gather information.To ensure fair energy dissipation between the nodes,LEACH randomly chooses cluster heads(CHs)and allocates this role to the various nodes based on a round-robin management mechanism.The intrusion-detection procedure was then utilized to determine whether intruders were present in the network.Within the WSN,a Fuzzy interference rule was utilized to distinguish the malicious nodes from legal nodes.Subsequently,an ANN was employed to distinguish the harmful nodes from suspicious nodes.The effectiveness of the proposed approach was validated using metrics that attained 97%accuracy,97%specificity,and 97%sensitivity of 95%.Thus,it was proved that the LEACH and Fuzzy-based IDS approaches are the best choices for securing data transmission in an energy-efficient manner.展开更多
高效准确的短期负荷预测是电力系统安全稳定与经济运行的重要保障。针对峰荷与谷荷预测误差较大的问题,提出一种基于栅格法提取负荷曲线特征的卷积神经网络和长短期记忆网络(convolutional neural network and long short term memory n...高效准确的短期负荷预测是电力系统安全稳定与经济运行的重要保障。针对峰荷与谷荷预测误差较大的问题,提出一种基于栅格法提取负荷曲线特征的卷积神经网络和长短期记忆网络(convolutional neural network and long short term memory network,CNN-LSTM)混合预测模型。首先,采用K-Medoids算法对日负荷曲线聚类,将各聚类中心作为典型代表日负荷曲线。采用栅格法将典型代表日负荷曲线划分为若干个区间并依次编号,提取负荷曲线的特征。然后,将各典型代表日负荷曲线特征与对应负荷类型历史数据重构成新的特征集输入到CNN-LSTM混合神经网络中。利用CNN挖掘数据间的特征形成新的特征向量,再将该特征向量输入到LSTM中进行预测。最后,以美国新英格兰地区2012至2013年电力负荷数据集为例进行仿真验证。结果表明,所提方法在不同日期下的负荷预测精度均有所提升,并且在提升日负荷平均预测精度的同时,有效提升了峰荷、谷荷的预测精度。展开更多
Finding clusters based on density represents a significant class of clustering algorithms.These methods can discover clusters of various shapes and sizes.The most studied algorithm in this class is theDensity-Based Sp...Finding clusters based on density represents a significant class of clustering algorithms.These methods can discover clusters of various shapes and sizes.The most studied algorithm in this class is theDensity-Based Spatial Clustering of Applications with Noise(DBSCAN).It identifies clusters by grouping the densely connected objects into one group and discarding the noise objects.It requires two input parameters:epsilon(fixed neighborhood radius)and MinPts(the lowest number of objects in epsilon).However,it can’t handle clusters of various densities since it uses a global value for epsilon.This article proposes an adaptation of the DBSCAN method so it can discover clusters of varied densities besides reducing the required number of input parameters to only one.Only user input in the proposed method is the MinPts.Epsilon on the other hand,is computed automatically based on statistical information of the dataset.The proposed method finds the core distance for each object in the dataset,takes the average of these distances as the first value of epsilon,and finds the clusters satisfying this density level.The remaining unclustered objects will be clustered using a new value of epsilon that equals the average core distances of unclustered objects.This process continues until all objects have been clustered or the remaining unclustered objects are less than 0.006 of the dataset’s size.The proposed method requires MinPts only as an input parameter because epsilon is computed from data.Benchmark datasets were used to evaluate the effectiveness of the proposed method that produced promising results.Practical experiments demonstrate that the outstanding ability of the proposed method to detect clusters of different densities even if there is no separation between them.The accuracy of the method ranges from 92%to 100%for the experimented datasets.展开更多
During the last three decades,evolutionary algorithms(EAs)have shown superiority in solving complex optimization problems,especially those with multiple objectives and non-differentiable landscapes.However,due to the ...During the last three decades,evolutionary algorithms(EAs)have shown superiority in solving complex optimization problems,especially those with multiple objectives and non-differentiable landscapes.However,due to the stochastic search strategies,the performance of most EAs deteriorates drastically when handling a large number of decision variables.To tackle the curse of dimensionality,this work proposes an efficient EA for solving super-large-scale multi-objective optimization problems with sparse optimal solutions.The proposed algorithm estimates the sparse distribution of optimal solutions by optimizing a binary vector for each solution,and provides a fast clustering method to highly reduce the dimensionality of the search space.More importantly,all the operations related to the decision variables only contain several matrix calculations,which can be directly accelerated by GPUs.While existing EAs are capable of handling fewer than 10000 real variables,the proposed algorithm is verified to be effective in handling 1000000 real variables.Furthermore,since the proposed algorithm handles the large number of variables via accelerated matrix calculations,its runtime can be reduced to less than 10%of the runtime of existing EAs.展开更多
磨音信号是反映磨机运行状态的一个重要参数,准确区分不同状态下的磨机信号将直接影响后续磨机运行参数优化的结果。通过聚类算法可以对磨音信号进行分类,为使磨音信号聚类效果更优,不仅需要类内距离小,还需要类间距离尽可能大。由此提...磨音信号是反映磨机运行状态的一个重要参数,准确区分不同状态下的磨机信号将直接影响后续磨机运行参数优化的结果。通过聚类算法可以对磨音信号进行分类,为使磨音信号聚类效果更优,不仅需要类内距离小,还需要类间距离尽可能大。由此提出一种基于局部稳定性加权动态时间规划(Local Stability Dynamic Time Warping,LSDTW)和优化k-medoids的磨音信号聚类方法。首先为克服动态时间规划(Dynamic Time Warping,DTW)得到的计算结果对噪声高度敏感的缺点,使用局部稳定性估计对DTW加权计算来降低噪声对计算结果的影响,其次针对k-medoids聚类需要多次计算才能确定聚类个数的不足,提出使用异常迭代模式(Abnormal Pattern,AP)优化k-medoids方法选取代表性的初始集群中心。采用优化k-medoids方法对LS-DTW的结果进行聚类分析,以平均轮廓系数作为评价标准,对比LS-DTW-k-medoids、DTW-k-medoids、DTW-优化k-medoids、k-means++算法效果可知,经本文方法聚类后,类内紧致性更优。展开更多
Various types of plasma events emerge in specific parameter ranges and exhibit similar characteristics in diagnostic signals,which can be applied to identify these events.A semisupervised machine learning algorithm,th...Various types of plasma events emerge in specific parameter ranges and exhibit similar characteristics in diagnostic signals,which can be applied to identify these events.A semisupervised machine learning algorithm,the k-means clustering algorithm,is utilized to investigate and identify plasma events in the J-TEXT plasma.This method can cluster diverse plasma events with homogeneous features,and then these events can be identified if given few manually labeled examples based on physical understanding.A survey of clustered events reveals that the k-means algorithm can make plasma events(rotating tearing mode,sawtooth oscillations,and locked mode)gathering in Euclidean space composed of multi-dimensional diagnostic data,like soft x-ray emission intensity,edge toroidal rotation velocity,the Mirnov signal amplitude and so on.Based on the cluster analysis results,an approximate analytical model is proposed to rapidly identify plasma events in the J-TEXT plasma.The cluster analysis method is conducive to data markers of massive diagnostic data.展开更多
As a mainstream research direction in the field of image segmentation,medical image segmentation plays a key role in the quantification of lesions,three-dimensional reconstruction,region of interest extraction and so ...As a mainstream research direction in the field of image segmentation,medical image segmentation plays a key role in the quantification of lesions,three-dimensional reconstruction,region of interest extraction and so on.Compared with natural images,medical images have a variety of modes.Besides,the emphasis of information which is conveyed by images of different modes is quite different.Because it is time-consuming and inefficient to manually segment medical images only by professional and experienced doctors.Therefore,large quantities of automated medical image segmentation methods have been developed.However,until now,researchers have not developed a universal method for all types of medical image segmentation.This paper reviews the literature on segmentation techniques that have produced major breakthroughs in recent years.Among the large quantities of medical image segmentation methods,this paper mainly discusses two categories of medical image segmentation methods.One is the improved strategies based on traditional clustering method.The other is the research progress of the improved image segmentation network structure model based on U-Net.The power of technology proves that the performance of the deep learning-based method is significantly better than that of the traditional method.This paper discussed both advantages and disadvantages of different algorithms and detailed how these methods can be used for the segmentation of lesions or other organs and tissues,as well as possible technical trends for future work.展开更多
文摘Cluster analysis is one of the major data analysis methods widely used for many practical applications in emerging areas of data mining. A good clustering method will produce high quality clusters with high intra-cluster similarity and low inter-cluster similarity. Clustering techniques are applied in different domains to predict future trends of available data and its uses for the real world. This research work is carried out to find the performance of two of the most delegated, partition based clustering algorithms namely k-Means and k-Medoids. A state of art analysis of these two algorithms is implemented and performance is analyzed based on their clustering result quality by means of its execution time and other components. Telecommunication data is the source data for this analysis. The connection oriented broadband data is given as input to find the clustering quality of the algorithms. Distance between the server locations and their connection is considered for clustering. Execution time for each algorithm is analyzed and the results are compared with one another. Results found in comparison study are satisfactory for the chosen application.
基金This research was funded by the National Natural Science Foundation of China(Grant No.72001190)by the Ministry of Education’s Humanities and Social Science Project via the China Ministry of Education(Grant No.20YJC630173)by Zhejiang A&F University(Grant No.2022LFR062).
文摘Data stream clustering is integral to contemporary big data applications.However,addressing the ongoing influx of data streams efficiently and accurately remains a primary challenge in current research.This paper aims to elevate the efficiency and precision of data stream clustering,leveraging the TEDA(Typicality and Eccentricity Data Analysis)algorithm as a foundation,we introduce improvements by integrating a nearest neighbor search algorithm to enhance both the efficiency and accuracy of the algorithm.The original TEDA algorithm,grounded in the concept of“Typicality and Eccentricity Data Analytics”,represents an evolving and recursive method that requires no prior knowledge.While the algorithm autonomously creates and merges clusters as new data arrives,its efficiency is significantly hindered by the need to traverse all existing clusters upon the arrival of further data.This work presents the NS-TEDA(Neighbor Search Based Typicality and Eccentricity Data Analysis)algorithm by incorporating a KD-Tree(K-Dimensional Tree)algorithm integrated with the Scapegoat Tree.Upon arrival,this ensures that new data points interact solely with clusters in very close proximity.This significantly enhances algorithm efficiency while preventing a single data point from joining too many clusters and mitigating the merging of clusters with high overlap to some extent.We apply the NS-TEDA algorithm to several well-known datasets,comparing its performance with other data stream clustering algorithms and the original TEDA algorithm.The results demonstrate that the proposed algorithm achieves higher accuracy,and its runtime exhibits almost linear dependence on the volume of data,making it more suitable for large-scale data stream analysis research.
基金sponsored by the National Natural Science Foundation of P.R.China(Nos.62102194 and 62102196)Six Talent Peaks Project of Jiangsu Province(No.RJFW-111)Postgraduate Research and Practice Innovation Program of Jiangsu Province(Nos.KYCX23_1087 and KYCX22_1027).
文摘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.
文摘Implementing machine learning algorithms in the non-conducive environment of the vehicular network requires some adaptations due to the high computational complexity of these algorithms.K-clustering algorithms are simplistic,with fast performance and relative accuracy.However,their implementation depends on the initial selection of clusters number(K),the initial clusters’centers,and the clustering metric.This paper investigated using Scott’s histogram formula to estimate the K number and the Link Expiration Time(LET)as a clustering metric.Realistic traffic flows were considered for three maps,namely Highway,Traffic Light junction,and Roundabout junction,to study the effect of road layout on estimating the K number.A fast version of the PAM algorithm was used for clustering with a modification to reduce time complexity.The Affinity propagation algorithm sets the baseline for the estimated K number,and the Medoid Silhouette method is used to quantify the clustering.OMNET++,Veins,and SUMO were used to simulate the traffic,while the related algorithms were implemented in Python.The Scott’s formula estimation of the K number only matched the baseline when the road layout was simple.Moreover,the clustering algorithm required one iteration on average to converge when used with LET.
基金supported in part by NUS startup grantthe National Natural Science Foundation of China (52076037)。
文摘Although many multi-view clustering(MVC) algorithms with acceptable performances have been presented, to the best of our knowledge, nearly all of them need to be fed with the correct number of clusters. In addition, these existing algorithms create only the hard and fuzzy partitions for multi-view objects,which are often located in highly-overlapping areas of multi-view feature space. The adoption of hard and fuzzy partition ignores the ambiguity and uncertainty in the assignment of objects, likely leading to performance degradation. To address these issues, we propose a novel sparse reconstructive multi-view evidential clustering algorithm(SRMVEC). Based on a sparse reconstructive procedure, SRMVEC learns a shared affinity matrix across views, and maps multi-view objects to a 2-dimensional humanreadable chart by calculating 2 newly defined mathematical metrics for each object. From this chart, users can detect the number of clusters and select several objects existing in the dataset as cluster centers. Then, SRMVEC derives a credal partition under the framework of evidence theory, improving the fault tolerance of clustering. Ablation studies show the benefits of adopting the sparse reconstructive procedure and evidence theory. Besides,SRMVEC delivers effectiveness on benchmark datasets by outperforming some state-of-the-art methods.
文摘Clustering a social network is a process of grouping social actors into clusters where intra-cluster similarities among actors are higher than inter-cluster similarities. Clustering approaches, i.e. , k-medoids or hierarchical, use the distance function to measure the dissimilarities among actors. These distance functions need to fulfill various properties, including the triangle inequality (TI). However, in some cases, the triangle inequality might be violated, impacting the quality of the resulting clusters. With experiments, this paper explains how TI violates while performing traditional clustering techniques: k-medoids, hierarchical, DENGRAPH, and spectral clustering on social networks and how the violation of TI affects the quality of the resulting clusters.
基金the Science and Technology Project of State Grid Corporation of China,Grant Number 5108-202304065A-1-1-ZN.
文摘Stochastic unit commitment is one of the most powerful methods to address uncertainty. However, the existingscenario clustering technique for stochastic unit commitment cannot accurately select representative scenarios,which threatens the robustness of stochastic unit commitment and hinders its application. This paper providesa stochastic unit commitment with dynamic scenario clustering based on multi-parametric programming andBenders decomposition. The stochastic unit commitment is solved via the Benders decomposition, which decouplesthe primal problem into the master problem and two types of subproblems. In the master problem, the committedgenerator is determined, while the feasibility and optimality of generator output are checked in these twosubproblems. Scenarios are dynamically clustered during the subproblem solution process through the multiparametric programming with respect to the solution of the master problem. In other words, multiple scenariosare clustered into several representative scenarios after the subproblem is solved, and the Benders cut obtainedby the representative scenario is generated for the master problem. Different from the conventional stochasticunit commitment, the proposed approach integrates scenario clustering into the Benders decomposition solutionprocess. Such a clustering approach could accurately cluster representative scenarios that have impacts on theunit commitment. The proposed method is tested on a 6-bus system and the modified IEEE 118-bus system.Numerical results illustrate the effectiveness of the proposed method in clustering scenarios. Compared withthe conventional clustering method, the proposed method can accurately select representative scenarios whilemitigating computational burden, thus guaranteeing the robustness of unit commitment.
文摘Multi-view Subspace Clustering (MVSC) emerges as an advanced clustering method, designed to integrate diverse views to uncover a common subspace, enhancing the accuracy and robustness of clustering results. The significance of low-rank prior in MVSC is emphasized, highlighting its role in capturing the global data structure across views for improved performance. However, it faces challenges with outlier sensitivity due to its reliance on the Frobenius norm for error measurement. Addressing this, our paper proposes a Low-Rank Multi-view Subspace Clustering Based on Sparse Regularization (LMVSC- Sparse) approach. Sparse regularization helps in selecting the most relevant features or views for clustering while ignoring irrelevant or noisy ones. This leads to a more efficient and effective representation of the data, improving the clustering accuracy and robustness, especially in the presence of outliers or noisy data. By incorporating sparse regularization, LMVSC-Sparse can effectively handle outlier sensitivity, which is a common challenge in traditional MVSC methods relying solely on low-rank priors. Then Alternating Direction Method of Multipliers (ADMM) algorithm is employed to solve the proposed optimization problems. Our comprehensive experiments demonstrate the efficiency and effectiveness of LMVSC-Sparse, offering a robust alternative to traditional MVSC methods.
基金This paper is supported by State Grid Gansu Electric Power Company Science and Technology Project(20220515003).
文摘To solve the problems of a few optical fibre line fault samples and the inefficiency of manual communication optical fibre fault diagnosis,this paper proposes a communication optical fibre fault diagnosis model based on variational modal decomposition(VMD),fuzzy entropy(FE)and fuzzy clustering(FC).Firstly,based on the OTDR curve data collected in the field,VMD is used to extract the different modal components(IMF)of the original signal and calculate the fuzzy entropy(FE)values of different components to characterize the subtle differences between them.The fuzzy entropy of each curve is used as the feature vector,which in turn constructs the communication optical fibre feature vector matrix,and the fuzzy clustering algorithm is used to achieve fault diagnosis of faulty optical fibre.The VMD-FE combination can extract subtle differences in features,and the fuzzy clustering algorithm does not require sample training.The experimental results show that the model in this paper has high accuracy and is relevant to the maintenance of communication optical fibre when compared with existing feature extraction models and traditional machine learning models.
文摘The k-means algorithm is a popular data clustering technique due to its speed and simplicity. However, it is susceptible to issues such as sensitivity to the chosen seeds, and inaccurate clusters due to poor initial seeds, particularly in complex datasets or datasets with non-spherical clusters. In this paper, a Comprehensive K-Means Clustering algorithm is presented, in which multiple trials of k-means are performed on a given dataset. The clustering results from each trial are transformed into a five-dimensional data point, containing the scope values of the x and y coordinates of the clusters along with the number of points within that cluster. A graph is then generated displaying the configuration of these points using Principal Component Analysis (PCA), from which we can observe and determine the common clustering patterns in the dataset. The robustness and strength of these patterns are then examined by observing the variance of the results of each trial, wherein a different subset of the data keeping a certain percentage of original data points is clustered. By aggregating information from multiple trials, we can distinguish clusters that consistently emerge across different runs from those that are more sensitive or unlikely, hence deriving more reliable conclusions about the underlying structure of complex datasets. Our experiments show that our algorithm is able to find the most common associations between different dimensions of data over multiple trials, often more accurately than other algorithms, as well as measure stability of these clusters, an ability that other k-means algorithms lack.
文摘Wireless sensor networks(WSN)gather information and sense information samples in a certain region and communicate these readings to a base station(BS).Energy efficiency is considered a major design issue in the WSNs,and can be addressed using clustering and routing techniques.Information is sent from the source to the BS via routing procedures.However,these routing protocols must ensure that packets are delivered securely,guaranteeing that neither adversaries nor unauthentic individuals have access to the sent information.Secure data transfer is intended to protect the data from illegal access,damage,or disruption.Thus,in the proposed model,secure data transmission is developed in an energy-effective manner.A low-energy adaptive clustering hierarchy(LEACH)is developed to efficiently transfer the data.For the intrusion detection systems(IDS),Fuzzy logic and artificial neural networks(ANNs)are proposed.Initially,the nodes were randomly placed in the network and initialized to gather information.To ensure fair energy dissipation between the nodes,LEACH randomly chooses cluster heads(CHs)and allocates this role to the various nodes based on a round-robin management mechanism.The intrusion-detection procedure was then utilized to determine whether intruders were present in the network.Within the WSN,a Fuzzy interference rule was utilized to distinguish the malicious nodes from legal nodes.Subsequently,an ANN was employed to distinguish the harmful nodes from suspicious nodes.The effectiveness of the proposed approach was validated using metrics that attained 97%accuracy,97%specificity,and 97%sensitivity of 95%.Thus,it was proved that the LEACH and Fuzzy-based IDS approaches are the best choices for securing data transmission in an energy-efficient manner.
文摘高效准确的短期负荷预测是电力系统安全稳定与经济运行的重要保障。针对峰荷与谷荷预测误差较大的问题,提出一种基于栅格法提取负荷曲线特征的卷积神经网络和长短期记忆网络(convolutional neural network and long short term memory network,CNN-LSTM)混合预测模型。首先,采用K-Medoids算法对日负荷曲线聚类,将各聚类中心作为典型代表日负荷曲线。采用栅格法将典型代表日负荷曲线划分为若干个区间并依次编号,提取负荷曲线的特征。然后,将各典型代表日负荷曲线特征与对应负荷类型历史数据重构成新的特征集输入到CNN-LSTM混合神经网络中。利用CNN挖掘数据间的特征形成新的特征向量,再将该特征向量输入到LSTM中进行预测。最后,以美国新英格兰地区2012至2013年电力负荷数据集为例进行仿真验证。结果表明,所提方法在不同日期下的负荷预测精度均有所提升,并且在提升日负荷平均预测精度的同时,有效提升了峰荷、谷荷的预测精度。
基金The author extends his appreciation to theDeputyship forResearch&Innovation,Ministry of Education in Saudi Arabia for funding this research work through the project number(IFPSAU-2021/01/17758).
文摘Finding clusters based on density represents a significant class of clustering algorithms.These methods can discover clusters of various shapes and sizes.The most studied algorithm in this class is theDensity-Based Spatial Clustering of Applications with Noise(DBSCAN).It identifies clusters by grouping the densely connected objects into one group and discarding the noise objects.It requires two input parameters:epsilon(fixed neighborhood radius)and MinPts(the lowest number of objects in epsilon).However,it can’t handle clusters of various densities since it uses a global value for epsilon.This article proposes an adaptation of the DBSCAN method so it can discover clusters of varied densities besides reducing the required number of input parameters to only one.Only user input in the proposed method is the MinPts.Epsilon on the other hand,is computed automatically based on statistical information of the dataset.The proposed method finds the core distance for each object in the dataset,takes the average of these distances as the first value of epsilon,and finds the clusters satisfying this density level.The remaining unclustered objects will be clustered using a new value of epsilon that equals the average core distances of unclustered objects.This process continues until all objects have been clustered or the remaining unclustered objects are less than 0.006 of the dataset’s size.The proposed method requires MinPts only as an input parameter because epsilon is computed from data.Benchmark datasets were used to evaluate the effectiveness of the proposed method that produced promising results.Practical experiments demonstrate that the outstanding ability of the proposed method to detect clusters of different densities even if there is no separation between them.The accuracy of the method ranges from 92%to 100%for the experimented datasets.
基金This work was supported in part by the National Key Research and Development Program of China(2018AAA0100100)the National Natural Science Foundation of China(61822301,61876123,61906001)+2 种基金the Collaborative Innovation Program of Universities in Anhui Province(GXXT-2020-051)the Hong Kong Scholars Program(XJ2019035)Anhui Provincial Natural Science Foundation(1908085QF271).
文摘During the last three decades,evolutionary algorithms(EAs)have shown superiority in solving complex optimization problems,especially those with multiple objectives and non-differentiable landscapes.However,due to the stochastic search strategies,the performance of most EAs deteriorates drastically when handling a large number of decision variables.To tackle the curse of dimensionality,this work proposes an efficient EA for solving super-large-scale multi-objective optimization problems with sparse optimal solutions.The proposed algorithm estimates the sparse distribution of optimal solutions by optimizing a binary vector for each solution,and provides a fast clustering method to highly reduce the dimensionality of the search space.More importantly,all the operations related to the decision variables only contain several matrix calculations,which can be directly accelerated by GPUs.While existing EAs are capable of handling fewer than 10000 real variables,the proposed algorithm is verified to be effective in handling 1000000 real variables.Furthermore,since the proposed algorithm handles the large number of variables via accelerated matrix calculations,its runtime can be reduced to less than 10%of the runtime of existing EAs.
文摘磨音信号是反映磨机运行状态的一个重要参数,准确区分不同状态下的磨机信号将直接影响后续磨机运行参数优化的结果。通过聚类算法可以对磨音信号进行分类,为使磨音信号聚类效果更优,不仅需要类内距离小,还需要类间距离尽可能大。由此提出一种基于局部稳定性加权动态时间规划(Local Stability Dynamic Time Warping,LSDTW)和优化k-medoids的磨音信号聚类方法。首先为克服动态时间规划(Dynamic Time Warping,DTW)得到的计算结果对噪声高度敏感的缺点,使用局部稳定性估计对DTW加权计算来降低噪声对计算结果的影响,其次针对k-medoids聚类需要多次计算才能确定聚类个数的不足,提出使用异常迭代模式(Abnormal Pattern,AP)优化k-medoids方法选取代表性的初始集群中心。采用优化k-medoids方法对LS-DTW的结果进行聚类分析,以平均轮廓系数作为评价标准,对比LS-DTW-k-medoids、DTW-k-medoids、DTW-优化k-medoids、k-means++算法效果可知,经本文方法聚类后,类内紧致性更优。
基金supported by the National Magnetic Confinement Fusion Science Program of China(Nos.2018YFE0301104 and 2018YFE0301100)National Natural Science Foundation of China(Nos.12075096 and 51821005)。
文摘Various types of plasma events emerge in specific parameter ranges and exhibit similar characteristics in diagnostic signals,which can be applied to identify these events.A semisupervised machine learning algorithm,the k-means clustering algorithm,is utilized to investigate and identify plasma events in the J-TEXT plasma.This method can cluster diverse plasma events with homogeneous features,and then these events can be identified if given few manually labeled examples based on physical understanding.A survey of clustered events reveals that the k-means algorithm can make plasma events(rotating tearing mode,sawtooth oscillations,and locked mode)gathering in Euclidean space composed of multi-dimensional diagnostic data,like soft x-ray emission intensity,edge toroidal rotation velocity,the Mirnov signal amplitude and so on.Based on the cluster analysis results,an approximate analytical model is proposed to rapidly identify plasma events in the J-TEXT plasma.The cluster analysis method is conducive to data markers of massive diagnostic data.
基金supported partly by the Open Project of State Key Laboratory of Millimeter Wave under Grant K202218partly by Innovation and Entrepreneurship Training Program of College Students under Grants 202210700006Y and 202210700005Z.
文摘As a mainstream research direction in the field of image segmentation,medical image segmentation plays a key role in the quantification of lesions,three-dimensional reconstruction,region of interest extraction and so on.Compared with natural images,medical images have a variety of modes.Besides,the emphasis of information which is conveyed by images of different modes is quite different.Because it is time-consuming and inefficient to manually segment medical images only by professional and experienced doctors.Therefore,large quantities of automated medical image segmentation methods have been developed.However,until now,researchers have not developed a universal method for all types of medical image segmentation.This paper reviews the literature on segmentation techniques that have produced major breakthroughs in recent years.Among the large quantities of medical image segmentation methods,this paper mainly discusses two categories of medical image segmentation methods.One is the improved strategies based on traditional clustering method.The other is the research progress of the improved image segmentation network structure model based on U-Net.The power of technology proves that the performance of the deep learning-based method is significantly better than that of the traditional method.This paper discussed both advantages and disadvantages of different algorithms and detailed how these methods can be used for the segmentation of lesions or other organs and tissues,as well as possible technical trends for future work.