检索结果-内蒙古大学图书馆

Unsupervised Competitive Learning clustering and Visual Method to Obtain Accurate Trajectories From Noisy Repetitive GPS Data

引用

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS 2025年第2期26卷 1562-1572页

作者： Mariotto, Flavio Tonioli Yoma, Nestor Becerra de Almeida, Madson Cortes Univ Estadual Campinas Fac Engn Elect & Computacao FEEC Dept Sistema Energia DSE BR-13083970 Campinas Brazil Univ Chile Dept Elect Engn Santiago 8370451 Chile

To make the proper planning of bus public transportation systems, especially with the introduction of electric buses to the fleets, it is essential to characterize the routes, patterns of traffic, speed, constraints, and presence of high slopes. Currently, GPS (Global Position System) is available worldwide in the fleet. However, they often produce datasets of poor quality, with low data rates, loss of information, noisy samples, and eventual paths not belonging to regular bus routes. Therefore, extracting useful information from these poor data is a challenging task. The current paper proposes a novel method based on an unsupervised competitive density clustering algorithm to obtain hot spot clusters of any density. The clusters are a result of their competition for the GPS samples. Each cluster attracts GPS samples until a maximum radius from its centroid and thereafter moves toward the most density areas. The winning clusters are sorted using a novel distance metric with the support of a visual interface, forming a sequence of points that outline the bus trajectory. Finally, indicators are correlated to the clusters making a trajectory characterization and allowing extensive assessments. According to the actual case studies, the method performs well with noisy GPS samples and the loss of information. The proposed method presents quite a fixed parameter, allowing fair performance for most GPS datasets without needing custom adjustments. It also proposes a framework for preparing the input GPS dataset, clustering, sorting the clusters to outline the trajectory, and making the trajectory characterization.

关键词： Global Positioning System Trajectory clustering algorithms Roads Public transportation Noise measurement Data mining Competitive learning Accuracy Visual analytics clustering methods machine learning competitive learning GPS trajectory analysis visual analytics

来源：评论

学校读者我要写书评

暂无评论

TRUNC: A Transfer Learning Unsupervised Network for Data clustering

引用

IEEE ACCESS 2025年 13卷 46282-46298页

作者： Xavier, Rita Peller, John de Castro, Leandro Nunes Univ Estadual Campinas Sch Technol BR-13484332 Limeira SP Brazil Florida Gulf Coast Univ Dendritic Human Ctr AI & Data Sci Inst Dept Comp & Software Engn Ft Myers FL 10501 USA

There is a demand for effective clustering methods, especially given the increasing complexity of data scenarios in modern applications. Motivated by the limitations of traditional clustering algorithms, particularly in handling noise, imbalanced data, and large-scale datasets, this work introduces TRUNC, a Transfer Learning Unsupervised Network for Data clustering. TRUNC leverages a bio-inspired approach, utilizing a single-layer feed-forward winner-takes-all neural network enhanced with transfer learning to optimize clustering performance. The proposed algorithm is evaluated across a range of synthetic and real-world datasets, demonstrating its robustness and performance over conventional and state-of-the-art clustering methods. Key contributions include the integration of transfer learning into the clustering process, a detailed sensitivity analysis of the TL parameter, and extensive experimentation that confirms the efficiency and generalization capability of TRUNC. Results indicate that TRUNC improves clustering quality and convergence speed, offering a competitive and scalable solution for various clustering tasks. This proposal opens new avenues for the integration of TRUNC's principles with other bio-inspired algorithms and its application across diverse domains.

关键词： Neurons Transfer learning Neural networks clustering algorithms Data models Vectors Biological system modeling Training Convergence Adaptation models Unsupervised learning self-organizing neural network transfer learning data clustering bio-inspired computing

来源：评论

学校读者我要写书评

暂无评论

Scalable Semi-Supervised clustering via Structural Entropy With Different Constraints

引用

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING 2025年第1期37卷 478-492页

作者： Zeng, Guangjie Peng, Hao Li, Angsheng Wu, Jia Liu, Chunyang Yu, Philip S. Beihang Univ State Key Lab Software Dev Environm Beijing 1000191 Peoples R China Macquarie Univ Dept Comp Sydney NSW 2109 Australia Didi Chuxing Beijing 100193 Peoples R China Univ Illinois Dept Comp Sci Chicago IL 60607 USA

Semi-supervised clustering leverages prior information in the form of constraints to achieve higher-quality clustering outcomes. However, most existing methods struggle with large-scale datasets owing to their high time and space complexity. Moreover, they encounter the challenge of seamlessly integrating various constraints, thereby limiting their applicability. In this paper, we present Scalable Semi-supervised clustering via Structural Entropy (SSSE), a novel method that tackles scalable datasets with different types of constraints from diverse sources to perform both semi-supervised partitioning and hierarchical clustering, which is fully explainable compared to deep learning-based methods. Specifically, we design objectives based on structural entropy, integrating constraints for semi-supervised partitioning and hierarchical clustering. To achieve scalability on data size, we develop efficient algorithms based on graph sampling to reduce the time and space complexity. To achieve generalization on constraint types, we formulate a uniform view for widely used pairwise and label constraints. Extensive experiments on real-world clustering datasets at different scales demonstrate the superiority of SSSE in clustering accuracy and scalability with different constraints. Additionally, Cell clustering experiments on single-cell RNA-seq datasets demonstrate the functionality of SSSE for biological data analysis.

关键词： Entropy Scalability Vegetation Encoding Data analysis clustering methods clustering algorithms Accuracy Complexity theory Similarity learning Biological data analysis scalable clustering semi-supervised clustering structural entropy

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Software Defect Prediction Through Multiview clustering

引用

IEEE TRANSACTIONS ON RELIABILITY 2025年

作者： Li, Zhiqiang Zhang, Hongyu Jing, Xiao-Yuan Yu, Wangyang Liu, Yueyue Shaanxi Normal Univ Sch Comp Sci Xian 710119 Peoples R China Chongqing Univ Sch Big Data & Software Engn Chongqing 401331 Peoples R China Wuhan Univ Sch Comp Sci Wuhan 430072 Peoples R China Guangdong Univ Petrochem Technol Guangdong Prov Key Lab Petrochem Equipment Fault D Maoming 525011 Peoples R China Univ Newcastle Sch Informat & Phys Sci Callaghan NSW 2308 Australia

The core goal of software defect prediction (SDP) is to identify modules with a high likelihood of defects, thereby enabling prioritization of quality assurance activities with low inspection effort. There are many supervised defect prediction models that are extensively studied. However, these methods require the need for labeling data to get enough training modules, which will cause a lot of waste of human resources. Cross-project defect prediction primarily reuses models trained on other projects with enough historical data. However, this strategy is often hindered by large distribution differences across different projects and privacy concerns of data. Unsupervised learning technique is an alternative solution to the unlabeled data, but it mainly focuses on single-view prediction by concatenating all the software metrics. This ignores the diversity and complementarity of different types of metrics. This study proposes a novel approach, namely, multiview unsupervised software defect prediction (MUSDP). It aims to collaboratively learn the diversity and complementarity of different views to build a robust and reliable defect prediction model. Extensive experiments on 28 releases from eight software projects indicate that MUSDP exhibits superior or comparable results regarding G-mean, AUC, P-opt, and Recall@20% compared to competing supervised and unsupervised methods. For the interpretation of MUSDP, the number of added and deleted lines significantly influence its predictions.

关键词： Software Predictive models Data models Software measurement Labeling Codes clustering algorithms Training data Training Supervised learning Interpretability multiview spectral clustering (SC) software defect prediction (SDP) unsupervised defect prediction (UDP)

来源：评论

学校读者我要写书评

暂无评论

LT-PEM Fuel Cells Diagnosis Based on EIS, clustering, and Automatic Parameter Selection

引用

IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY 2025年第2期74卷 2513-2526页

作者： Chanal, Damien Steiner, Nadia Yousfi Chamagne, Didier Pera, Marie-Cecile Univ Franche Comte FEMTO ST Inst CNRS FCLAB F-90000 Belfort France

In the field of fuel cells, early detection of faulty conditions can significantly improve the lifetime. Then, signal analysis techniques such as electrochemical impedance spectroscopy combined with machine learning algorithms can generate a representation of the system state of health space using data in known conditions. Although the onboard measurement of EIS can be done by controlling the harmonic content of the power converter at the output of the fuel cell, the implementation of in-vehicle diagnostic algorithms is still limited by the absence of a large database listing the evolution of performance throughout the life cycle. This paper presents a fast-diagnostic method able to consider the occurrence of new data to adapt the dimensional space representing the health state and compensate for the lack of data. Available measurements come from two low-temperature proton exchange membrane fuel cell technologies characterized by two laboratories. The results presented in the paper show that the automatic parameter selection provides performances as good as the ones obtained by an expert. The feasibility of the approach has also been demonstrated on a low-cost embedded platform.

关键词： Fuel cells Feature extraction Databases Standardization clustering algorithms Impedance Hydrogen clustering methods fault diagnosis feature selection fuel cells standardization

来源：评论

学校读者我要写书评

暂无评论

Deep Spectral clustering With Projected Adaptive Feature Selection

引用

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 2025年 63卷

作者： Zhao, Yang Bi, Zixuan Zhu, Peican Yuan, Aihong Li, Xuelong Northwestern Polytech Univ OPt & Elect iOPEN Sch Artificial Intelligence Xian 710072 Peoples R China Shanghai Artificial Intelligence Lab Shanghai 200232 Peoples R China Northwest A&F Univ Coll Informat Engn Yangling 712100 Peoples R China China Telecom Corp Ltd Inst Artificial Intelligence Tele AI Beijing 100033 Peoples R China

In the past era of explosive data growth, how to deal with large-scale, unlabeled remote sensing images (RSIs) has become a concern. Due to the lack of data labels, unsupervised methods are usually used to deal with them. As one of the best unsupervised algorithms, spectral clustering (SC) is a very effective data processing and analysis technology. However, the scalability of SC affects its development in the era of big data. Deep network technology has developed rapidly in the past, and it is often used to solve the problem of out-of-sample expansion (OOSE) that traditional machine learning cannot solve. However, because RSI is usually high-dimensional data, it is easy to cause dimension explosion in deep networks. The main motivation of this work is to solve the above problems. We have designed a new algorithm called deep SC with projected adaptive feature selection (DSCFS), which benefits from deep learning theory and feature projection technology. We use a neural network to map RSI data and then further process the data through regularization embedding. At the same time, adaptive feature projection is applied to extract the main features of RSI, and the loss feedback network is calculated through these two steps. A large number of experiments show that the performance of our proposed method is better than other mainstream methods.

关键词： Feature extraction clustering algorithms Artificial intelligence Electronic mail Unsupervised learning Optics Remote sensing Laplace equations High dimensional data Autoencoders Deep network feature projection remote sensing images (RSIs)

来源：评论

学校读者我要写书评

暂无评论

A Topological-Indicators-Based k-Means clustering Algorithm and Its Application in Time Series Data: A Case Study on Sea Level Variability in Peninsular Malaysia

引用

IEEE ACCESS 2025年 13卷 46514-46533页

作者： Lin, Zixin Zulkepli, Nur Fariha Syaqina Kasihmuddin, Mohd Shareduwan Bin Mohd Gobithaasan, Rudrusamyr Univ Sains Malaysia USM Sch Math Sci George Town 11800 Penang Malaysia

Traditional k-means clustering is widely used to analyze regional and temporal variations in time series data, such as sea levels. However, its accuracy can be affected by limitations, particularly when applied to datasets with mixed groups or significant noise. In this study, we analyzed monthly sea level data derived from daily time series at 14 tide gauge stations along the coastline of Peninsular Malaysia. To enhance traditional k-means clustering, we propose a hybrid approach that combines clustering techniques with topological data analysis (TDA). Specifically, we integrate k-means and its variant, k-means++, with persistent homology, the primary tool in TDA, to capture topological insights from the datasets. The proposed approach clusters the 14 tide gauge stations based on predefined topological features, and the probability of data points from each station belonging to specific clusters is computed. The results demonstrate that our approach significantly improves the performance of traditional k-means clustering by incorporating topological information, compared to using clustering without such insights.

关键词： Sea level Sea measurements Point cloud compression Time series analysis Vectors Noise clustering algorithms Biological system modeling Analytical models Tides Machine learning computational topology environmental sustainability sea level variability

来源：评论

学校读者我要写书评

暂无评论

Sparse Multi-View K-Means clustering

引用

IEEE ACCESS 2025年 13卷 46773-46793页

作者： Yang, Miin-Shen Parveen, Shazia Chung Yuan Christian Univ Dept Appl Math Taoyuan 32023 Taiwan

In machine learning, k-means clustering is an unsupervised leaning technique to partition the data into k clusters that are homogeneous within the cluster and heterogeneous between clusters. The k-means algorithm assigns equal importance to all features in a single view. Several multi-view clustering techniques have been developed over the last few decades, including multi-view k-means (MVKM). In MVKM, patterns can be detected more accurately, and clustering performance is improved by applying each view individually for pattern discovery. The demand for highly accurate and flexible methods in machine learning and artificial intelligence has increased the use of multi-view datasets. These datasets are complicated and pose challenges for researchers in real-life situations. A novel type of sparse k-means clustering algorithm, sparse MVKM (S-MVKM), is proposed in this study for multi-view data. The main aim of our proposed S-MVKM algorithm is to achieve sparsity in features and shrink unimportant features to exactly zero using Lasso (least absolute shrinkage and selection operator). For considering theoretic behaviors of the S-MVKM algorithm, we create a convergence theorem to ensure that any convergent subsequence in S-MVKM will ultimately tend to an optimal solution. The proposed S-MVKM algorithm is then compared with several existing algorithms, such as traditional MVKM, feature reduction MVKM (FR-MVKM), MVKM with adaptive sparse membership weight allocation (M-VASM), and redundant and sparse feature learning (RSF-MVKM). Based on the three clustering performance measures of accuracy rate, Rand index, and normalized mutual information, we make the comparisons between them by using several synthetic and real-world datasets. The comparative results demonstrate that the proposed S-MVKM clustering algorithm effectively surpasses these competing clustering algorithms in terms of these three performance measures. Finally, we make concluding remarks that provide potential g

关键词： clustering algorithms Linear programming Machine learning algorithms Convergence Accuracy Data mining Social networking (online) Partitioning algorithms Internet of Things Industries K-means multi-view k-means (MVKM) sparsity sparse MVKM (S-MVKM) feature reduction performance measures

来源：评论

学校读者我要写书评

暂无评论

Enhanced Vector Quantization for Embedded Machine Learning: A Post-Training Approach With Incremental clustering

引用

IEEE ACCESS 2025年 13卷 17440-17456页

作者： Flores, Thommas K. S. Medeiros, Morsinaldo Silva, Marianne Costa, Daniel G. Silva, Ivanovitch Univ Fed Rio Grande do Norte PPgEEC BR-59078970 Natal Brazil Univ Fed Alagoas SI BR-57200000 Penedo Brazil Univ Porto Fac Engn SYSTEC ARISE P-4200465 Porto Portugal

TinyML enables the deployment of Machine Learning (ML) models on resource-constrained devices, addressing a growing need for efficient, low-power AI solutions. However, significant challenges remain due to strict memory, processing, and energy limitations. This study introduces a novel method to optimize Post-Training Quantization (PTQ), a widely used technique for reducing model size, by integrating Vector Quantization (VQ) with incremental clustering. While VQ is a technique that reduces model size by grouping similar parameters, incremental clustering, implemented via the AutoCloud K-Fixed algorithm, preserves accuracy during compression. This combined approach was validated on an automotive dataset predicting CO2 emissions from vehicle sensor measurements such as mass air flow, intake pressure, temperature, and speed. The model was quantized and deployed on Macchina A0 hardware, demonstrating over 90% compression with negligible accuracy loss. Results show improved performance and deployment efficiency, showcasing the potential of this combined technique for real-world embedded applications.

关键词： clustering algorithms Computational modeling Accuracy Adaptation models Training Embedded systems Computational efficiency Automotive engineering Vector quantization Data models post-training quantization vector quantization automotive sensors pollution

来源：评论

学校读者我要写书评

暂无评论

Assessing clustering methods using Shannon's entropy

引用

INFORMATION SCIENCES 2025年 689卷

作者： Hoayek, Anis Rulliere, Didier Univ Clermont Auvergne Inst Henri Fayol CNRS Mines St EtienneUMR 6158 LIMOS F-42023 St Etienne France

Unsupervised clustering techniques are crucial for effectively partitioning datasets into meaningful subgroups. In this paper, we introduce a novel clustering fuzziness metric based on Shannon's entropy, which quantifies the level of uncertainty in clustering assignments. This metric is employed to develop a statistical test for evaluating clustering algorithms and to propose parametric corrections to enhance their performance. We provide empirical evidence showing that our approach significantly improves clustering quality compared to well-known algorithms such as Fuzzy K-Means (FKM) and Fanny. For instance, applying our correction method led to a notable decrease in Jensen-Shannon Divergence (JSD) values, indicating enhanced clustering accuracy. Our methodology is demonstrated through comprehensive experiments on both simulated and real-world datasets. Additionally, our approach proves effective in determining the optimal number of clusters, showing superiority over traditional methods. These results highlight the practical benefits of our metric and correction techniques for improving clustering performance.

关键词： clustering algorithms

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：