检索结果-内蒙古大学图书馆

11th International Conference on E-Health and Bioengineering (EHB)

作者： Zhdanov, Aleksei Bulev, Daniil Dolganov, Anton Kulyabin, Mikhail Ural Fed Univ Engn Sch Informat Technol Telecommun & Control Sy Ekaterinburg Russia Univ Erlangen Nurnberg Dept Comp Sci Pattern Recognit Lab Erlangen Germany

ISBN: (纸本)9783031625015;9783031625022

Electroretinogram (ERG) signals are commonly used in electrophysiological research to measure the retina's electrical responses to light stimuli. Using publicly available signals from the IEEE DataPort repository, 11 clustering algorithms were assessed, including traditional and modern methods, to categorize the ERG signals into distinct groups. The clustering results were evaluated using metrics such as the Silhouette Coefficient, Calinski-Harabasz score, and Davies-Bouldin score. The study found that the Affinity Propagation algorithm was the most effective approach in classifying ERG signals. This method uses a messagepassing framework to update the examples and responsibilities of each data point based on a damping factor and the affinities between the data points. The study's findings can contribute to enhancing the classification of ERG signals, ultimately improving the accuracy of electrophysiological research.

关键词： Electroretinogram Machine Learning clustering algorithms Ophthalmology Signal Processing

来源：评论

学校读者我要写书评

暂无评论

Augmenting Energy Sustainability of Static Nodes Using Hybrid KGNN-AHP Driven Approach for IoT-Based Heterogeneous WSN

引用

IEEE ACCESS 2025年 13卷 3320-3354页

作者： Blessina Preethi, R. Saranya Nair, M. Vellore Inst Technol Chennai 600127 India

As the nodes used in Internet of Things (IoT)- based wireless sensor network (WSN) are constrained by the limited source of energy, contemporary applications incorporate the heterogeneous energy model WSN. Although the node energies are heterogeneous in application, further study is required to improve the potential of heterogeneous WSN and not much is explored on Graph Neural Network (GNN) clustering and Routing methods, which impels this study on novel optimal clustering of nodes with efficient cluster head selection and routing for heterogeneous WSN. In the proposed work, a novel clustering model with K-nearest neighbour (KNN) and GNN clustering, followed by the Analytic Hierarchical Process (AHP) weight-based cluster head (CH) selection method to find the optimal and eligible CH with classified fitness functions. The chosen parameters for the fitness functions are the base station to node distance, lifetime of the nodes in the network, average number of neighbouring nodes in the cluster, peak power transmission by the node in the network, and the average lifetime of nodes in the cluster. Additionally, the proposed algorithm ensures the eligibility of the optimal relay nodes using the GNNSage based routing and steady-state of data transmission from source nodes to the base station. As the model is designed for static nodes deployment for monitoring environmental changes in realtime application. The simulation results exhibit the enhancement of network lifetime and data transmission by following the proposed algorithm for static network.

关键词： Wireless sensor networks Protocols Routing Batteries clustering algorithms Graph neural networks Energy efficiency Base stations Energy consumption Adaptation models Analytic hierarchical process graph neural network GraphSAGE heterogeneous WSN Internet of Things

来源：评论

学校读者我要写书评

暂无评论

COMPARATIVE ASSESSMENT OF clustering STRATEGY FOR REDUCING COMPUTATIONAL LOAD IN DISTRIBUTED FUSION TRANSFORMER (DAFTA)

COMPARATIVE ASSESSMENT OF CLUSTERING STRATEGY FOR REDUCING C...

引用

IEEE International Geoscience and Remote Sensing Symposium (IGARSS)

作者： Shahare, Vivek Banolia, Chaman Deshpande, Shailesh Indian Inst Technol Dharwad Karnataka India TCS Res TRDDC Pune Maharashtra India

ISBN: (纸本)9798350360332;9798350360325

Hyperspectral image (HSI) classification is widely used for many critical remote sensing applications that require high accuracy. Their classification is a challenging task due to the large size of HSI datasets. One approach to reduce the size of HSI datasets is to exploit patch similarity. Pixels with the same class tend to exhibit similar features in space, so similar patches are unlikely to provide any new information for classification. (Distributed Architecture for Fusion Transformer training Acceleration) DAFTA is a multi-modal data fusion transformer that integrates information from multiple modalities and sources, such as HSI and LiDAR, to improve classification accuracy. This work proposes a method for the DAFTA framework to reduce the size of HSI datasets with less clustering time than the existing approach. We use a clustering approach to identify and eliminate similar patches that belong to the same class with the least amount of time. The experimental results show that the proposed method can significantly reduce the size of the training dataset in minimum time without sacrificing classification accuracy, which enhances the efficiency of distributed training.

关键词： remote sensing hyperspectral data fusion transformers distributed training clustering algorithms DAFTA

来源：评论

学校读者我要写书评

暂无评论

Assessment of K-Means Algorithm to Evaluate Nearshore Wave Climate

引用

IEEE JOURNAL OF OCEANIC ENGINEERING 2025年第2期50卷 583-593页

作者： Castro, Elisa Iuppa, Claudio Musumeci, Rosaria Ester Xibilia, Maria Gabriella Patane, Luca Foti, Enrico Cavallaro, Luca Univ Catania Dept Civil Engn & Architecture I-95123 Catania Italy Univ Messina Dept Engn I-98166 Contrada Di Dio Italy

Accidents near ports have increased due to the ongoing expansion of maritime trade. These accidents have various causes, including adverse weather conditions. Accurate wave climate forecasts can help mitigate the risks associated with marine accidents. While numerical models are commonly used for obtaining nearshore wave climate forecasts, their high computational cost makes them impractical for wave climate forecasting and nowcasting. Artificial neural networks (ANNs) offer a potential solution to this limitation. However, existing ANNs have primarily focused on specific single points within the study areas, such as piers and port entrances. Enhancing early-warning strategies requires a broader understanding of the wave climate across larger areas. Thorough examinations of extensive areas with varying physical attributes can result in significant computational time requirements. The main objective of this study is to evaluate a clustering technique able to identify homogeneous areas to improve future applications of ANNs to assess nearshore wave characteristics in actual situations. The area around the port of Augusta (Sicily), one of the most important ports in Italy, serves as a case study in this article. Results show an optimal performance by applying the clustering algorithm K-means, capable of capturing the wave climate characteristics of the study area.

关键词： Meteorology clustering algorithms Seaports Sea measurements Accidents Numerical models Partitioning algorithms K-means maritime accidents SWAN wave climate

来源：评论

学校读者我要写书评

暂无评论

Improving Cyber Defense With DNS Query clustering Analysis 11

Improving Cyber Defense With DNS Query Clustering Analysis

引用

11th International Conference on Wireless Networks and Mobile Communications (WINCOM) - Inclusive and Intelligent Connectivity

作者： Khaoula, Radi Imane, Moughit Mohamed, Moughit Sultan Moulay Slimane Univ Natl Sch Appl Sci LaSTI Lab Khouribga Morocco Sultan Moulay Slimane Univ Natl Sch Appl Sci LIPIM Lab Khouribga Morocco

ISBN: (纸本)9798350377873;9798350377866

The Domain Name System (DNS) serves as a critical component of internet infrastructure, facilitating the translation of domain names to IP addresses. However, its widespread usage also renders it a prominent target for cyber threats. Detecting and mitigating malicious activities within DNS queries are imperative for upholding online security standards. This study investigates the effectiveness of clustering algorithms in identifying and analyzing malicious activity within domain names and IP addresses. By employing various clustering techniques such as kmeans, DBSCAN, hierarchical clustering, and Gaussian mixture models (GMM) on DNS query datasets, insightful patterns and trends are uncovered to reveal potential threats. To augment the accuracy and efficiency of clustering algorithms, advanced preprocessing techniques like principal component analysis (PCA) and t-distributed stochastic neighbor embedding (t-SNE) are utilized. These techniques aid in reducing data dimensionality while preserving essential characteristics, thereby enabling the capture of intricate relationships within DNS queries and the more effective identification of malicious behavior. Our research underscores the superiority of specific clustering algorithms in accurately detecting malicious activity compared to others. Furthermore, we demonstrate how integrating dimensionality reduction techniques significantly enhances the performance of these algorithms, enabling more precise identification and classification of threats. By offering a comprehensive evaluation of clustering algorithms and preprocessing methods for DNS query analysis, this study equips cybersecurity professionals with valuable insights to safeguard against potential threats and fortify the security of DNS infrastructure. The findings presented in this paper provide compelling evidence of the efficacy of clustering techniques in combating cyber attacks, paving the way for advanced defense mechanisms and improved cybersecurity pra

关键词： DNS malicious activity clustering algorithms k-means DBSCAN hierarchical clustering GMM preprocessing techniques PCA t-SNE.

来源：评论

学校读者我要写书评

暂无评论

Tensor Representation-Based Multiview Graph Contrastive Learning for IoE Intelligence

引用

IEEE INTERNET OF THINGS JOURNAL 2025年第4期12卷 3482-3492页

作者： Gao, Yuan Zhao, Qian Yang, Laurence T. Yang, Jing Yang, Jieming Zhengzhou Univ Sch Comp Sci & Artificial Intelligence Zhengzhou 570001 Peoples R China Hainan Univ Sch Comp Sci & Technol Haikou 570100 Peoples R China Huazhong Univ Sci & Technol Sch Comp Sci & Technol Wuhan 430074 Peoples R China St Francis Xavier Univ Dept Comp Sci Antigonish NS B2G 2W5 Canada

As a prevalent computing paradigm, graph computing provides an effective service strategy for the analysis of big data within the Internet of Everything (IoE), particularly for clustering graph-structured data in the IoE. Among various methods, graph contrastive clustering, which is devoted to revealing the intrinsic structure of graphs and efficiently grouping nodes into distinct clusters by contrasting positive-negative counterparts, has attracted widely attention in recent years. However, the existing methods seriously ignore the graph topology and node attributes when setting positive and negative sample pairs, which further leads to node semantic inconsistency. To this end, we design a novel tensor representation-based multiview contrastive graph representation learning framework, including adaptive data augmentation, high-confidence sample pairs construction, and a simple yet effective self-optimizing module guided by clustering objective function, to address issues of graph contrastive learning in ignoring complementary information among the topology and attributes. Specifically, by jointly modeling the graph structure and multiview node attributes, the new proposed clustering model can concurrently mine both hard positive and negative samples, and dynamically enhance the weight allocation of the hard samples during the learning process. Then leveraging the characteristics of graph-structured data, we incorporate a small subset of nodes with the highest similarity as additional positive samples to improve the discriminative power of the proposed model. Furthermore, a self-optimizing clustering module is introduced to enhance the algorithm's performance. Experimental results on four commonly used IoE-related data sets validate that our proposed approach can achieve state-of-the-art clustering performance.

关键词： Tensors Internet of Things Representation learning clustering algorithms Task analysis Topology clustering methods Graph clustering graph representation learning tensor computing

来源：评论

学校读者我要写书评

暂无评论

A Characteristics-Based Least Common Multiple Algorithm to Optimize Magnetic-Field-Based Indoor Localization

引用

IEEE INTERNET OF THINGS JOURNAL 2025年第8期12卷 10463-10478页

作者： Rafique, Hamaad Patti, Davide Palesi, Maurizio La Delfa, Gaetano Carmelo Univ Catania Dept Elect Elect & Comp Engn I-95125 Catania Italy

clustering is an unsupervised learning technique that groups data based on similarity criteria. Traditional methods like K-Means and agglomerative clustering often require predefined parameters, struggle with irregular cluster shapes, and fail to classify subcluster points in magnetic fingerprint-based indoor localization. This study proposes the characteristics-based least common multiple (LCM) algorithm to address these challenges. This novel approach autonomously determines cluster number and shape while accurately classifying misclassified points based on characteristic similarities using LCM. We evaluated the proposed technique using state-of-the-art metrics and tested it in magnetic-field-based indoor localization scenarios. Comparisons were made with real-time and benchmark datasets, alongside traditional clustering methods. Results demonstrate that LCM significantly enhances localization accuracy, achieving a mean absolute error rate of 0.1 m.

关键词： Location awareness clustering algorithms Shape Fingerprint recognition Internet of Things Accuracy Wireless fidelity Real-time systems clustering methods Time series analysis Automatic clustering data clustering indoor localization least common multiple (LCM) machine learning novel clustering technique

来源：评论

学校读者我要写书评

暂无评论

Collaborative clustering with Heterogeneous algorithms

Collaborative Clustering with Heterogeneous Algorithms

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Sublime, Jeremie Grozavu, Nistor Bennani, Younes Cornuejols, Antoine AgroParisTech UMR INRA MIA 518 16 Rue Claude Bernard F-75231 Paris 5 France Univ Paris 13 Sorbonne Paris Cite UPN UMR CNRS 7030 F-93430 Villetaneuse France

ISBN: (纸本)9781479919598

The aim of collaborative clustering is to reveal the common underlying structures found by different algorithms while analyzing data. The fundamental concept of collaboration is that the clustering algorithms operate locally but collaborate by exchanging information about the local structures found by each algorithm. In this framework, the one purpose of this article is to introduce a new method which allows to reinforce the clustering process by exchanging information between several results acquired by different clustering algorithms. The originality of our proposed approach is that the collaboration step can use clustering results obtained from any type of algorithm during the local phase. This article gives the theoretical foundations of our approach as well as some experimental results. The proposed approach has been validated on several data sets and the results have shown to be very competitive.

关键词： clustering algorithms clustering process information exchange algorithms Dataset Cluster Analysis

来源：评论

学校读者我要写书评

暂无评论

Research on Enterprise Digital Economy Management System Based on clustering Algorithm 24

Research on Enterprise Digital Economy Management System Bas...

引用

4th International Conference on Internet of Things and Machine Learning

作者： Liu, Lin Taishan Univ Sch Management & Econ Tai An Shandong Peoples R China

ISBN: (纸本)9798400710353

With the rapid advancement of the global digital economy, enterprise management systems are undergoing profound transformations. Traditional management models no longer suffice to meet modern enterprises' demands for efficiency and intelligence;consequently, data-driven management tools have increasingly become a focal point of research. clustering algorithms, as a core technology in data analysis, enhance the precision and efficiency of enterprise management through data segmentation and analysis. This study investigates the application of clustering algorithms in enterprise digital economy management systems, particularly focusing on the refinement of the K-means algorithm and its parallel processing implementation. By optimizing initial clustering centers and incorporating Lagrange multipliers and attribute weighting strategies, this paper proposes a more adaptive enterprise management solution, applied practically in modules such as employee management, financial management, and supplier management. Experimental results demonstrate that the improved algorithm significantly enhances both clustering effectiveness and computational efficiency, providing robust theoretical and technical support for digital enterprise management.

关键词： clustering algorithms digital economy enterprise management systems

来源：评论

学校读者我要写书评

暂无评论

Modeling and Verification of MRSCAN Based on MapReduce Framework

引用

IEEE TRANSACTIONS ON RELIABILITY 2025年

作者： Zuo, Zhengkang Ke, Yuhan Hu, Ying Huang, Qing Zeng, Zhicheng Wang, Changjing Jiangxi Normal Univ Sch Comp & Informat Engn Nanchang 330022 Peoples R China Jiangxi Normal Univ Jiangxi Prov Key Lab High Performance Comp Nanchang 330022 Peoples R China Jiangxi Normal Univ State Int Sci & Technol Cooperat Base Networked Su Nanchang 330022 Peoples R China Jiangxi Normal Univ Language Intelligence Res Ctr Nanchang 330022 Peoples R China East China Univ Technol Sch Software Fuzhou 344099 Peoples R China

Network clustering (graph clustering) plays a crucial role in discovering the inherent structures within networks. MapReduce-based structural clustering algorithm for networks (MRSCANs) designed based on MapReduce's parallel computing model, efficiently handles large-scale data. However, MRSCAN can only be tested through experiments, and its correctness cannot be guaranteed. To address this issue, this article has achieved the first implementation of functional MRSCAN modeling and subjected it to rigorous mechanized verification in Isabelle. First, based on Google's MapReduce model type definition and higher-order generic functions, a general MapReduce-based algorithm functional modeling framework is constructed across the fundamental global phases of Map, Shuffle, and Reduce. Moreover, diverse strategies are devised during the Shuffle phases according to user requirements, enhancing the applicability and generality of the MapReduce functional modeling framework. Second, formalizing the definition of MRSCAN, is delineated into four key steps: similarity calculation, core calculation, dimension expansion, and structural clustering. Furthermore, the MapReduce functional modeling framework is applied to these four steps to achieve the functional modeling of MRSCAN, which improves efficiency compared to other structural clustering algorithm for networks (SCANs) algorithms. Lastly, a verification framework for MapReduce-based algorithms is proposed at both the global and shuffle stages. Based on this framework, the correctness and reliability of MRSCAN are ensured. The model framework and verification framework of MapReduce-based algorithms proposed in this article can not only address functional modeling and verification of MRSCAN but also provide a reference for a series of other MapReduce-based functional program designs and proofs.

关键词： clustering algorithms Computational modeling Reliability Programming Scalability Internet Software algorithms Loss measurement Logic Complexity theory Function algorithm MapReduce-based structural clustering algorithm for networks (MRSCANs) modeling framework proof in Isabelle verification framework

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：