检索结果-内蒙古大学图书馆

STICA: Spanning-Tree-Based Identification of Clusters With Agglomeration for Numerical Data

IEEE ACCESS 2024年 12卷 150430-150450页

作者： Khan, Mozammel H. A. East West Univ Dept Comp Sci & Engn Dhaka 1212 Bangladesh

clustering is an unsupervised method used to group data points based on their similarity. It has many applications in data mining and preprocessing. algorithms used for arbitrarily shaped clustering have several real-world applications. However, most of these algorithms require input parameters, such as the number of clusters, neighborhood radius, density threshold, and number of grids. However, it is difficult to estimate these parameters. A pragmatic approach is to develop a simple but parameterless clustering algorithm. To achieve this goal, we propose a parameterless algorithm for arbitrarily shaped clustering called spanning-tree-based identification of clusters (STIC) for numerical data. We consider data points as nodes of spanning trees (STs). We determined the STs' edges based on the Euclidean distances between the node pairs and an internally estimated edge weight threshold. We generated clusters using breadth-first search (BFS) based on the notion of a spanning-tree-based representation of a cluster. Data points that were not visited by the BFS process were outliers. This algorithm is effective for datasets with inter-cluster separation. The STIC algorithm divides heterogeneous clusters into sub-clusters. In this case, we performed agglomerative merging of sub-clusters and outliers based on the single-linkage distances (SLDs) of the clusters and an internally estimated SLD threshold. This extended algorithm is called the STIC with agglomeration (STICA). Our experimental results with artificial datasets demonstrate that the STICA algorithm outperforms several widely used distance-based algorithms for arbitrarily shaped clusters. However, for real-world datasets, the STICA algorithm behaves similarly to other distance-based algorithms.

关键词： clustering algorithms clustering methods Merging Training Standards Noise measurement Optics Distributed databases Accuracy Anomaly detection Arbitrarily shaped clustering parameterless clustering spanning-tree-based clustering cluster merging outlier detection

来源：评论

学校读者我要写书评

暂无评论

Distributed Matching-By-Clone Hungarian-Based Algorithm for Task Allocation of Multiagent Systems

引用

IEEE TRANSACTIONS ON ROBOTICS 2024年 40卷 851-863页

作者： Samiei, Arezoo Sun, Liang Klipsch Sch Elect & Comp Engn Las Cruces NM 88003 USA Dept Mech & Aerosp Engn Las Cruces NM 88003 USA

In this article, we present a novel approach, namely distributed matching-by-clone hungarian-based algorithm (DMCHBA), to multiagent task-allocation problems, in which the number of agents is smaller than the number of tasks. The proposed DMCHBA assumes that agents employ an implicit coordination mechanism and consists of two iterative phases, i.e., the communication phase and the assignment phase. In the communication phase, agents communicate with their connected neighbors and exchange their local knowledge base until they converge on the global knowledge base. In the assignment phase, each agent builds a squared cost matrix by cloning agents and adding pseudotasks when necessary, and applying the Hungarian method for task allocation. A local planning algorithm is then applied to identify the order of task execution for an agent. The proposed DMCHBA is proven to produce conflict-free assignments among agents in finite time. We compare the performance of DMCHBA with the consensus-based bundle algorithm, the distributed recursive Hungarian-based algorithms, and the cluster-based Hungarian algorithm (CBHA) in Monte-Carlo simulations with different numbers of agents and tasks. The numerical results reveal the superior convergence and optimality of DMCHBA over all other selected algorithms.

关键词： Task analysis Resource management clustering algorithms Costs Robots Multitasking Robot kinematics Autonomous robots autonomous systems distributed task allocation multiagent systems uncrewed systems

来源：评论

学校读者我要写书评

暂无评论

Big Data Optimization clustering Algorithm for Power Grid CPS Based on PSO 2

Big Data Optimization Clustering Algorithm for Power Grid CP...

引用

2nd International Conference on Data Science and Information System, ICDSIS 2024

作者： Yue, Junying Arimuzha Inner Mongolia Normal University College Of Computer Science And Technology Hohhot China

ISBN: (纸本)9798350381665

As power systems become increasingly intelligent and internet of things technology finds widespread use, the power grid cyber-physical system (CPS) has become pivotal in real-time monitoring, data processing, and intelligent control. Nonetheless, the generation of large-scale, high-dimensional real-time data within the power grid CPS system poses challenges to traditional data processing and analysis methods. In order to meet this challenge, this paper proposes a PSO-based power grid CPS big data optimization clustering algorithm. By introducing PSO algorithm, the algorithm makes full use of its global search and adaptive characteristics to cluster the massive data in the CPS system of power grid efficiently. Compared with the traditional clustering algorithm, the algorithm proposed in this paper is superior in mining data relations and finding abnormal situations. In the experimental verification stage, we use different scale and complexity power grid data sets to verify the feasibility and performance of the algorithm. The experimental results show that the power grid CPS big data optimization clustering algorithm based on PSO has achieved significant performance improvement in multiple data scenarios. The research results of this paper have important theoretical and practical significance for intelligent management, data mining and anomaly detection of power grid system. © 2024 IEEE.

关键词： clustering algorithms

来源：评论

学校读者我要写书评

暂无评论

Multiview Subspace clustering by an Enhanced Tensor Nuclear Norm

引用

IEEE TRANSACTIONS ON CYBERNETICS 2022年第9期52卷 8962-8975页

作者： Xia, Wei Zhang, Xiangdong Gao, Quanxue Shu, Xiaochuang Han, Jungong Gao, Xinbo Xidian Univ State Key Lab Integrated Serv Networks Xian 710071 Peoples R China Aberystwyth Univ Comp Sci Dept Aberystwyth SY23 3FL Dyfed Wales Chongqing Univ Posts & Telecommun Chongqing Key Lab Image Cognit Chongqing 400065 Peoples R China

Despite the promising preliminary results, tensor-singular value decomposition (t-SVD)-based multiview subspace is incapable of dealing with real problems, such as noise and illumination changes. The major reason is that tensor-nuclear norm minimization (TNNM) used in t-SVD regularizes each singular value equally, which does not make sense in matrix completion and coefficient matrix learning. In this case, the singular values represent different perspectives and should be treated differently. To well exploit the significant difference between singular values, we study the weighted tensor Schatten p-norm based on t-SVD and develop an efficient algorithm to solve the weighted tensor Schatten p-norm minimization (WTSNM) problem. After that, applying WTSNM to learn the coefficient matrix in multiview subspace clustering, we present a novel multiview clustering method by integrating coefficient matrix learning and spectral clustering into a unified framework. The learned coefficient matrix well exploits both the cluster structure and high-order information embedded in multiview views. The extensive experiments indicate the efficiency of our method in six metrics.

关键词： Tensors Minimization clustering methods Periodic structures Matrix decomposition Correlation clustering algorithms Multiview clustering spectral clustering tensor-singular value decomposition (t-SVD) weighted nuclear norm

来源：评论

学校读者我要写书评

暂无评论

Interrupted Sampling Repeater Jamming Recognition and Suppression Method Based on DBSCAN Algorithm

引用

IEEE ACCESS 2024年 12卷 62597-62608页

作者： Zhong, Qi He, Zhiyi Wei, Shaoren Beijing Inst Remote Sensing Equipment Beijing 100854 Peoples R China

The interrupted sampling repeater jamming (ISRJ) is a type of coherent jamming that can not only improve the detection threshold of adjacent range cells but also induce the radar system to track a false target. In order to suppress ISRJ, we have developed an interference identification and suppression method based on a clustering algorithm. First, we obtain the time-frequency (TF) domain matrix by applying the short-time Fourier transform (STFT) to the received signal containing the target echo and ISRJ. Then, the time-frequency domain matrix is subjected to threshold decision to obtain a series of over-threshold points which are then processed by the density-based spatial clustering of applications with noise (DBSCAN) algorithm to form a series of clusters. By analyzing the number of points of these clusters, we can identify whether the cluster corresponds to the target echo or ISRJ. Recognition mainly depends on the number of cluster points and the corresponding point threshold allowing for good recognition without the need for accurate estimation of ISRJ parameters. An accurate time-frequency domain filter is constructed based on the recognition results to avoid damaging the spectral structure of the target echo while suppressing the interference. We simulate the method and verify the effect of different combinations of input signal to noise ratio (SNR) and input signal to jamming ratio (SJR) on the performance of the method through Monte Carlo simulation. The simulation results demonstrate that this method has a good suppression effect for the three ISRJ modes under various combinations of SNR and SJR.

关键词： Time-frequency analysis Radar Jamming Radar tracking Interference suppression clustering algorithms Training Sampling methods Chirp modulation Fourier transforms Interrupted sampling repeater jamming (ISRJ) interference suppression linear frequency modulation (LFM) waveform density-based spatial clustering of applications with noise (DBSCAN) short-time Fourier transform (STFT)

来源：评论

学校读者我要写书评

暂无评论

Clustered and Thinned Antenna Array for Adaptive Interference Suppression via K-Medoids of Desired-Signal Steering Vector

引用

IEEE TRANSACTIONS ON ANTENNAS AND PROPAGATION 2024年第3期72卷 2867-2872页

作者： Lee, Yinman Natl Chi Nan Univ Dept Elect Engn Puli 54561 Taiwan

Both clustering and thinning are effective ways to reduce the implementation complexity of large-size antenna arrays. In this communication, we propose simple but efficient approaches to cluster and thin the antenna array for adaptive interference suppression. First, antenna elements are grouped using the $K$ -Medoids algorithm according to the knowledge of the desired-signal steering vector. Based on the clustered result, the minimum-variance distortionless response (MVDR) criterion with reduced dimension is then applied to ensure good adaptive-beamforming performance. To lower the implementation complexity and power consumption further, selected antenna elements are turned off on the basis of the perception from clustering. Simulation results show that under suitable settings, the adaptive interference-suppression capability of the proposed clustered and thinned antenna array can be consistently close to that of the optimal MVDR beamformer, and some reasonable compromises between the performance and both the required hardware and computational complexity are achieved.

关键词： Antenna arrays Adaptive arrays Computational complexity clustering algorithms Signal to noise ratio Hardware Costs Adaptive interference suppression clustered array K-medoids algorithm minimum-variance distortionless response (MVDR) thinned array

来源：评论

学校读者我要写书评

暂无评论

An Identification Method for Anomaly Types of Active Distribution Network Based on Data Mining

引用

IEEE TRANSACTIONS ON POWER SYSTEMS 2024年第4期39卷 5548-5560页

作者： Wang, Shaorui Lu, Tianguang Hao, Ran Li, Jing Guo, Yu He, Xueqian Han, Xueshan Shandong Univ Sch Elect Engn Jinan 250061 Peoples R China Harvard Univ Sch Engn & Appl Sci Cambridge MA 02138 USA Harvard Univ Harvard China Project Cambridge MA 02138 USA State Grid Corp China Natl Power Dispatching & Control Ctr Beijing 100000 Peoples R China

With the increasing penetration of distributed generators (DGs) and the growing demand for reliable power sources, it has become imperative to promptly identify anomalies in active distribution networks (ADNs). Additionally, anomaly identification is also crucial to assisting in targeting maintenance to handle failures after the action of relay protections. This study presents a highly accurate and rapidly responsive approach to identifying anomalies in ADNs. The proposed method uses three layers of intricately interconnected modules. In the first module, an iterative clustering layer is devised to process the original data. In the second module, an association rule layer is designed to identify hidden patterns based on the types of anomalies and the post-fault voltage. In the third module, a regression training layer is employed to train accurate identification models. The case study utilizes data samples collected by an IEEE 33-bus system with distributed generations as well as post-fault data from a practical distribution network. The results of comparative tests demonstrate that the proposed data-mining structure achieves reliable performance with high identification accuracy and short reaction times, indicating its ability to be applied to the state awareness system of ADNs.

关键词： Data mining clustering algorithms Reliability Training Distribution networks Load modeling Data models Active distribution network anomaly identification data mining distributed generator linear regression

来源：评论

学校读者我要写书评

暂无评论

Discovering Temporal Patterns for Event Sequence clustering via Policy Mixture Model

引用

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING 2022年第2期34卷 573-586页

作者： Wu, Weichang Yan, Junchi Yang, Xiaokang Zha, Hongyuan Shanghai Jiao Tong Univ AI Inst MoE Key Lab Artificial Intelligence Dept Elect Engn Shanghai 200240 Peoples R China Shanghai Jiao Tong Univ AI Inst MoE Key Lab Artificial Intelligence Dept Comp Sci & Engn Shanghai 200240 Peoples R China Shanghai Jiao Tong Univ AI Inst MoE Key Lab Artificial Intelligence Shanghai 200240 Peoples R China Georgia Inst Technol Sch Computat Sci & Engn Coll Comp Atlanta GA 30332 USA Shanghai Jiao Tong Univ Shanghai 200240 Peoples R China

Temporal point process (TPP) is an expressive tool for modeling the temporal pattern of event sequences. However, discovering temporal patterns for event sequences clustering is rarely studied in TPP modeling. To solve this problem, we take a reinforcement learning view whereby the observed sequences are assumed to be generated from a mixture of latent policies. The purpose is to cluster the sequences with different temporal patterns into the underlying policies while learning each of the policy model. The flexibility of our model lies in: i) all the components are networks including the policy network for modeling the temporal point process;ii) to handle varying-length event sequences, we resort to inverse reinforcement learning by decomposing the observed sequence into states (RNN hidden embedding of history) and actions (time interval to next event) in order to learn a reward function, it helps to achieve better performance or increasing efficiency compared to existing methods using rewards over the entire sequence such as log-likelihood or Wasserstein distance. We adopt an Expectation-Maximization algorithm, in E-step estimating the cluster labels for each sequence, in M-step aiming to learn the respective policy. Extensive experiments on synthetic and real-world datasets show the efficacy of our method against the state-of-the-arts.

关键词： Data models clustering algorithms Machine learning Stochastic processes Predictive models Earthquakes Social network services Time-series analysis temporal point processes reinforcement learning model-based clustering

来源：评论

学校读者我要写书评

暂无评论

Estimation of Marine Oil Spill Surface Characteristics Based on Laser-Induced Fluorescence With IA-DBSCAN

引用

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT 2024年 73卷 1页

作者： Sun, Qiankun Liu, Weifeng Wen, Chenglin Cai, Lei Shaanxi Univ Sci & Technol Sch Elect & Control Engn Xian 710021 Peoples R China Guangdong Univ Petrochem Technol Sch Automat Maoming 525000 Peoples R China Henan Inst Sci & Technol Sch Artificial Intelligence Xinxiang 453003 Peoples R China

Detection of marine oil spills and accurate estimation of the oil spill concentration distribution are essential in an emergency response. This article proposes an iterative adaptation density-based spatial clustering of applications with noise (IA-DBSCAN) algorithm, which aims to estimate the surface characteristics of marine oil spills, including source localization and concentration distribution. This algorithm addresses the issues of traditional density-based spatial clustering of applications with noise (DBSCAN) algorithms, such as the inability to obtain density information and the lack of autonomy in parameter selection. We utilize laser-induced fluorescence (LIF) technology to monitor marine oil spills. According to the Bernoulli model, we convert the collected fluorescence intensity into the density of fluorescent points. The density distribution of these fluorescent points can be used to estimate the characteristics of the oil spill surface. This brings other challenges: clustering under mixed-density conditions and accurately calculating the concentration of spilled oil. To address these challenges, a mean filtering method based on the distance-weighted from oil source points was proposed to smooth the concentration of fluorescent points, which improves the estimation accuracy of the oil spill surface as well as reduces the fluorescence detection errors caused by wave and environmental factors. Finally, tests were conducted in both marine simulation environments and laboratory simulation environments. The results of the tests indicate that the proposed method is capable of accurately estimating the characteristics of the oil spill surface.

关键词： Oils Sea surface clustering algorithms Fluorescence Estimation Surface emitting lasers Surface waves Concentration distribution laser-induced fluorescence (LIF) marine oil spill oil spill source location parameter estimation

来源：评论

学校读者我要写书评

暂无评论

Proactive Recommendation of Composite Services in Multi-Access Edge Computing

引用

IEEE TRANSACTIONS ON SERVICES COMPUTING 2024年第2期17卷 631-644页

作者： Liu, Zhizhong Sheng, Quan Z. Chu, Dianhui Xu, Xiaofei Zheng, Hedan Feng, Kai Yantai Univ Sch Comp & Control Engn Yantai 264005 Shandong Peoples R China Macquarie Univ Sch Comp Sydney NSW 2109 Australia Harbin Inst Technol Coll Comp Sci & Technol Weihai 264209 Shandong Peoples R China Henan Polytech Univ Coll Comp Sci & Technol Jiaozuo 454000 Henan Peoples R China

Multi-Access Edge Computing (MEC) is an emerging computing paradigm that brings services from centralized cloud to nearby network edge to improve users' Quality of Experience (QoE). With massive services from different domains being emerging in MEC, various powerful composite services can be created with simple services to satisfy users' complex needs. However, existing service composition methods follow a passive service model and cannot proactively recommend optimal composite services to users in MEC, which seriously affect users' service experience. To tackle this issue, we propose an approach for proactive recommendation of composite services based on demand prediction. Our approach consists of three steps. First, we predict a user's service demand based on an attention enhanced deep interaction network (AEDIN) model trained with clustered data. Then, we create the optimal composite service to satisfy the predicted demand with a mobility-aware services composition method, and finally, we proactively recommend the optimal composite service to the user. The extensive experiments have been carried out to verify our proposed approach and prove its performance superiority.

关键词： Predictive models Quality of service Context-aware services Hidden Markov models clustering algorithms Prediction algorithms Quality of experience Multi-access edge computing composite service recommendation service demand prediction deep learning

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：