检索结果-内蒙古大学图书馆

IEEE SSD International Multi-Conference on Systems, Signals and Devices

作者： Jaouher Chrouta Abderrahmen Zaafouri Mohamed Jemli Monitoring and Safety of Systems C3S ENSIT Research Unit on Control

Fuzzy c-means (FCM) and Gustafson-Kessel (GK) algorithms are the best popular fuzzy clustering techniques in terms of efficient, straightforward, and easy to implement. However, these algrithms are sensitive to initialization and easy to trap in the local minimum. The important issue is how to avoid getting a bad local minimum value to improve the cluster accuracy. In fact, the particle swarm algorithm is strong global searching ability which is based on swarm operation, it doesn't easily get into the local minimum and has a fast convergence speed. In order to overcome the weakness of traditional clustering algorithms and takes advantage of PSO, we integrate FCM and GK algorithms with fuzzy particle swarm algorithm (FCM-PSO and GK-PSO algorithms). In this paper, hybrid fuzzy clustering algorithms based on FCM, GK and PSO called FCM-PSO and GK-PSO are presented. A comparative study between the clustering algorithms is investigated to identify the parameter of irrigation station. Experimental results applied to the irrigation station show that the GK-PSO algorithm is more effective and robust compared to the other algorithms.

关键词： clustering algorithms Particle swarm optimization Partitioning algorithms Covariance matrices Convergence Irrigation Nonlinear systems

来源：评论

学校读者我要写书评

暂无评论

CNCMSA-ERCP: An Innovative Energy-Efficient clustering Routing Protocol for Improving the Performance of Industrial IoT

引用

IEEE INTERNET OF THINGS JOURNAL 2025年第9期12卷 11827-11840页

作者： Liu, Xin Cao, Qike Jin, Bo Zhou, Peng Beijing Univ Technol Beijing 100124 Peoples R China Henan Inst Sci & Technol Xinxiang 453003 Henan Peoples R China Univ Coimbra Inst Syst & Robot Dept Elect & Comp Engn P-3030290 Coimbra Portugal Portuguese Natl Acad Sci P-1600477 Lisbon Portugal Xinjiang Coll Sci & Technol Sch Informat Sci & Engn Korla 841000 Xinjiang Peoples R China Yanshan Univ Sch Informat Sci & Engn Qinhuangdao 066004 Peoples R China

As Industry 4.0 rapidly advances, optimizing energy and performance in industrial wireless sensor networks (IWSNs) for the Industrial Internet of Things (IIoT) is critical. Addressing the NP-hard challenge of IWSN clustering, this article introduces a novel multiobjective model to balance energy consumption and enhance service quality. Based on this model, a new energy-saving cluster routing protocol called CNCMSA-ECRP was designed. The protocol comprehensively considers four key indicators: 1) total residual energy;2) average network delay;3) average network packet loss rate;and 4) average distance from cluster heads (CHs) to the base station (BS), with the goal of achieving overall optimization of network performance. The CNCMSA-ECRP protocol employs our newly proposed chaotic niching clone moth swarm algorithm (CNCMSA). The algorithm introduces chaotic mapping operators in the initialization phase and clonal and niching operators in the evolution phase, enhancing the search capability and diversity of the algorithm, as well as improving clustering efficiency and network stability. Comparative experiments with four existing advanced clustering schemes-LEACH-C, EEHCHR, OptGACHE, and ESCVAD-demonstrate that CNCMSA-ECRP achieves at least a 16.38% increase in network lifespan and significantly reduces average data transmission delay and packet loss rate by at least 1.84% and 17.76%, respectively. This research provides a novel solution to the clustering routing problem in IWSN.

关键词： clustering algorithms Wireless sensor networks Industrial Internet of Things Quality of service Optimization Energy efficiency Routing Routing protocols Energy consumption Delays Chaotic niching clone cluster routing protocol industrial wireless sensor network (IWSN) moth swarm algorithm Quality of Service (QoS)

来源：评论

学校读者我要写书评

暂无评论

Survey on clustering Problems Using Metaheuristic algorithms

Survey on Clustering Problems Using Metaheuristic Algorithms

引用

International Conference on Reliability, Infocom Technologies and Optimization (ICRITO) (Trends and Future Directions)

作者： Shubhada Agarwal Sumit Kumar Amity University Noida Uttar Pradesh

ISBN: (数字)9798350350357

ISBN: (纸本)9798350350364

Data clustering is a highly sought-after approach to classify data objects. Different conventional partitioning and hierarchical clustering approaches are not capable of handling abstract and diversified real-world data generated in today's time. clustering is an optimization and NP-hard problem. The research in the last decade shows that metaheuristics are successful and reliable techniques to generate optimized solutions for problems having infinite search space. This paper explores the scope, feasibility, and related work of metaheuristics to solve the clustering problem. Keywords: clustering, Partial Swarm, Neighbourhood search.

关键词： Surveys NP-hard problem Metaheuristics clustering algorithms Search problems Market research Polynomials

来源：评论

学校读者我要写书评

暂无评论

Proactive Frequency Stability Scheme Based on Bayesian Filters and Spectral clustering

IEEE OPEN ACCESS JOURNAL OF POWER AND ENERGY

引用

IEEE OPEN ACCESS JOURNAL OF POWER AND ENERGY 2025年 12卷 100-110页

作者： Paramo, Gian Baquedano-Aguilar, Mario D. Bretas, Arturo Meyn, Sean Univ Florida Dept Elect & Comp Engn Gainesville FL 32611 USA Sandia Natl Labs Elect Grid Secur Albuquerque NM 87185 USA Univ Grenoble Alpes Grenoble INP CNRS G2Elab F-38000 Grenoble France

This work presents a proactive distributed model for power system frequency stability. High-level penetration of renewable energy sources into the grid have introduced unforeseen and unmodeled system dynamics. Underfrequency load shedding state-of-the-art solutions are reactive in design, with efficiency constrained by the modeling error. Being able to detect unstable conditions early makes it possible to generate optimized corrective actions. In this work, phasor measurement units are used to predict frequency values. When a disturbance is detected, the state of frequency is predicted a few seconds into the future via a particle filter algorithm. Corrective actions are modeled through a mixed integer linear programming algorithm within system areas established through spectral clustering. The solution is implemented on Matlab, considering IEEE test systems. The proactive design of the method combined with its multiple layers of optimization deliver results that outperform state-of-the-art solutions. Easy-to-implement model, without hard-to-derive parameters, highlights potential aspects towards real-life implementation.

关键词： Mathematical models Power system dynamics Particle measurements Filters clustering algorithms Atmospheric measurements Power system stability Phasor measurement units Weight measurement Time-frequency analysis spectral clustering dynamic state estimation mixed-integer linear programming particle filter frequency stability

来源：评论

学校读者我要写书评

暂无评论

Simple One-Step Multi-View clustering With Fast Similarity and Cluster Structure Learning

引用

IEEE SIGNAL PROCESSING LETTERS 2025年 32卷 1850-1854页

作者： Xia, Xianxian Huang, Dong Yang, Chen-Min He, Chaobo Wang, Chang-Dong South China Agr Univ Coll Math & Informat Guangzhou 510642 Peoples R China South China Normal Univ Sch Comp Sci Guangzhou 510635 Peoples R China Sun Yat Sen Univ Sch Comp Sci & Engn Guangzhou 510006 Peoples R China Guangdong Key Lab Big Data Anal & Proc Guangzhou 510006 Peoples R China

Multi-view clustering (MVC) is essential for integrating heterogeneous data from multiple sources. However, many existing approaches are hindered by high computational complexity and the separate optimization of similarity and cluster structures. In light of these challenges, this paper presents a novel anchor-based MVC method termed simple one-step multi-view clustering with fast similarity and cluster structure learning (SONIC), which models adaptive anchor learning, multi-view similarity structure learning, and discrete cluster structure learning in a joint framework. In particular, we employ the anchor-based multi-view similarity learning to capture the consensus manifold structure latent in multiple views, thereby constructing a unified bipartite graph with adaptive anchor learning and view weighting. Then we impose a low-rank constraint on the bipartite graph structure to directly reveal the desired number of clusters without additional post-processing. An efficient alternating minimization algorithm is developed to optimize the model, resulting in a computational complexity that scales linearly with the number of samples. Extensive experiments on eight benchmark datasets demonstrate the superior performance of SONIC in both clustering quality and computational efficiency.

关键词： Optimization Bipartite graph Similarity learning Linear programming Adaptation models Vectors Signal processing algorithms Laplace equations clustering algorithms Training Consensus information data clustering lare-scale clustering linear time multi-view clustering

来源：评论

学校读者我要写书评

暂无评论

Galaxy clustering and Classification using Machine Learning algorithms and XAI

Galaxy Clustering and Classification using Machine Learning ...

引用

International Conference on Information Technology Research (ICITR)

作者： Amasha Elvitigala U dani Navaratne Samadhi Rathnayake Kapila Dissanayaka Dept. of Information Technology Sri Lanka Institute of Information Technology Malabe Sri Lanka Dept. of Information Technology Sri Lanka Institute of Information Technology Kandy Sri Lanka Dept. of Information Technology Sri Lanka Institute of Information Technology Colombo Sri Lanka

ISBN: (数字)9798331511425

ISBN: (纸本)9798331511432

This paper presents a novel approach to galaxy clustering and analysis using machine learning (ML), coupled with Explainable Artificial Intelligence (XAI) methods. Traditional astronomical research often relies on manual analysis of galaxy clusters, which can be time-consuming and potentially biased. Our study aims to automate this process using advanced computational methods, focusing on clustering algorithms such as Gaussian Mixture Models (GMM) and Density-Based Spatial clustering of Applications with Noise (DBSCAN). We also employ classification algorithms including Random Forests, Decision Trees, and Support Vector Machines (SVM). The application of XAI techniques enhances the interpretability and trustworthiness of our models' outputs. By leveraging data from the Sloan Digital Sky Survey (SDSS), we demonstrate the efficacy of our approach in identifying and analyzing galaxy clusters, potentially revolutionizing the field of astronomical research.

关键词： Surveys Support vector machines Accuracy Machine learning algorithms Explainable AI Noise clustering algorithms Classification algorithms Random forests Gaussian mixture model

来源：评论

学校读者我要写书评

暂无评论

Comparing K-Means and DBSCAN algorithms for clustering Poverty Levels in Papua Islands

Comparing K-Means and DBSCAN Algorithms for Clustering Pover...

引用

Information Technology and Digital Applications (ICITDA), IEEE International Conference on

作者： Katherine Vanessa Irene Anindaputri Iswanto Katrina Wijaya Muhammad Fadlan Hidayat Computer Science Department School of Computer Science Bina Nusantara University Jakarta Indonesia

ISBN: (数字)9798331510732

ISBN: (纸本)9798331510749

Poverty has become one of the main challenges in Indonesia, especially in the eastern regions. According to data from the Central Bureau of Statistics (BPS), Papua Island currently ranks as the poorest province. As of 2018, the poverty rate in Papua was 27.6%, and as of March 2023, it was 26.03%. The government has made various efforts to address the issue of poverty in Papua, one of which is providing financial assistance in the form of BLT (Direct Cash Assistance). However, to achieve sustainable development, more effective and targeted strategies are needed. The government's efforts to address poverty in Papua must be supported by accurate and efficient data analysis. This study involves clustering data on impoverished populations from 42 districts/cities in Papua and West Papua, obtained from the Central Bureau of Statistics (BPS) website in 2023. The study aims to compare the clustering results using the K-Means and DBSCAN algorithms. The performance of these algorithms is determined by the highest Silhouette Index (SI) value. After the data underwent a standardization process, the clustering results using the K-Means algorithm achieved the best SI value of 0.6359 with K=4. Meanwhile, the clustering results using the DBSCAN algorithm achieved the best SI value of 0.3678 with eps 1.8 and MinPts 8. The SI value of the K-Means algorithm is higher than that of the DBSCAN algorithm, indicating that clustering using the K-Means algorithm is superior to DBSCAN.

关键词： Shape Government Noise Education clustering algorithms Standardization Indexes Resource management Sustainable development Information technology

来源：评论

学校读者我要写书评

暂无评论

Machine Learning algorithms to find out the best images from the pool of blurred and unoriented images and clustering based on similarities

Machine Learning algorithms to find out the best images from...

引用

Emerging Innovations and Advanced Computing (INNOCOMP), International Conference on

作者： Hetan Hemant Nandre Aniket Prashant More Gaurav Narendra Patil Gaurav Rajendra Patil Pushkar Joglekar Computer Engineering Vishwakarma Institute of Technology Pune India

ISBN: (数字)9798350376470

ISBN: (纸本)9798350376487

In the era of digital photography, the ease of capturing moments has reached unprecedented levels, rendering the creation of lasting memories a seemingly effortless task. However, not all images encapsulate the essence of cherished moments with the desired perfection. This innovative project proposes a groundbreaking solution to streamline the curation process, ensuring that only the most exceptional images are preserved in the vast sea of digital snapshots. The primary focus of this endeavor is to eliminate suboptimal captures, filtering out images plagued by blurriness or poor orientation. Furthermore, the project aims to organize these carefully curated images into coherent clusters based on their inherent similarities, creating a seamless and intuitive structure within photo collections. The central objective revolves around elevating the overall quality of photo collections by establishing a rigorous selection process. By harnessing the power of advanced techniques, such as image analysis algorithms and machine learning models, the project aims to automate the identification of optimal images. This not only enhances the efficiency of the curation process but also saves users valuable time and effort that would otherwise be spent manually sifting through a multitude of images. Through the application of cutting-edge technology, the project seeks to go beyond mere selection and extend its capabilities to categorizing the chosen images into organized clusters based on shared characteristics. This multifaceted approach ensures that users are not only presented with the best individual photos but also benefit from a well-organized and intelligently categorized photo collection. In summary, this project represents a significant leap forward in the realm of digital photography, combining innovation, efficiency, and technology to revolutionize the way we curate and preserve our most treasured memories. With a focus on automating the identification of optimal images and

关键词： Image quality Technological innovation Machine learning algorithms Filtering Memory management clustering algorithms Machine learning

来源：评论

学校读者我要写书评

暂无评论

Impact of Varying Distance-Based Fingerprint Similarity Metrics on Affinity Propagation clustering Performance in Received Signal Strength-Based Fingerprint Databases

IEEE OPEN JOURNAL OF SIGNAL PROCESSING

引用

IEEE OPEN JOURNAL OF SIGNAL PROCESSING 2024年 5卷 1005-1014页

作者： Yaro, Abdulmalik Shehu Maly, Filip Maly, Karel Prazak, Pavel Univ Hradec Kralove Fac Informat & Management Dept Informat & Quantitat Methods Hradec Kralove 50003 Czech Republic Ahmadu Bello Univ Dept Elect & Telecommun Engn Zaria 810106 Nigeria

The affinity propagation clustering (APC) algorithm is popular for fingerprint database clustering because it can cluster without pre-defining the number of clusters. However, the clustering performance of the APC algorithm heavily depends on the chosen fingerprint similarity metric, with distance-based metrics being the most commonly used. Despite its popularity, the APC algorithm lacks comprehensive research on how distance-based metrics affect clustering performance. This emphasizes the need for a better understanding of how these metrics influence its clustering performance, particularly in fingerprint databases. This paper investigates the impact of various distance-based fingerprint similarity metrics on the clustering performance of the APC algorithm. It identifies the best fingerprint similarity metric for optimal clustering performance for a given fingerprint database. The analysis is conducted across five experimentally generated online fingerprint databases, utilizing seven distance-based metrics: Euclidean, squared Euclidean, Manhattan, Spearman, cosine, Canberra, and Chebyshev distances. Using the silhouette score as the performance metric, the simulation results indicate that structural characteristics of the fingerprint database, such as the distribution of fingerprint vectors, play a key role in selecting the best fingerprint similarity metric. However, Euclidean and Manhattan distances are generally the preferable choices for use as fingerprint similarity metrics for the APC algorithm across most fingerprint databases, regardless of their structural characteristics. It is recommended that other factors, such as computational intensity and the presence or absence of outliers, be considered alongside the structural characteristics of the fingerprint database when choosing the appropriate fingerprint similarity metric for maximum clustering performance.

关键词： Fingerprint recognition Measurement clustering algorithms Databases Vectors Signal processing algorithms Location awareness Affinity propagation clustering distance-based similarity metrics fingerprint clustering silhouette score

来源：评论

学校读者我要写书评

暂无评论

Prototype Imputation Guided Incomplete Multi-View clustering

引用

IEEE SIGNAL PROCESSING LETTERS 2025年 32卷 1565-1569页

作者： Hong, Rui Chen, Xiao-Ping Zhou, Yan Liu, Hui Wan, Tiancai Univ Elect Sci & Technol China Chengdu 611731 Peoples R China Avic Chengdu Aircraft Design & Res Inst Chengdu 610091 Peoples R China

In real-world data collection, certain views may become partially unavailable due to sensor malfunctions or equipment faults, giving rise to the challenge of incomplete multi-view clustering (IMVC). Although some IMVC methods that utilize prototype completion yield acceptable performance, the majority implicitly assume the correct alignment of prototypes across all views. However, the generated prototypes could have the potential misalignment of prototypes across views. To tackle this issue, we propose a new Prototype Imputation Guided Incomplete Multi-view clustering (PIGIMC), which can reduce the impact of prototype mismatch. Specifically, RPCIC first employs a multi-view contrastive learning module to obtain consistent feature representations across various views. To mitigate the adverse impact caused by multi-view prototype mismatch, we propose a prototype matching learning to achieve prototype alignment. Lastly, we propose a prototype imputation strategy to impute the missing data, thereby achieving IMVC. Extensive experiments confirm that PIGIMC surpasses 11 state-of-the-art IMVC methods.

关键词： Prototypes Imputation Decoding Contrastive learning Training Data mining Machine learning Interference clustering algorithms Bipartite graph Incomplete multi-view clustering prototype alignment prototype matching 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：