检索结果-内蒙古大学图书馆

Machine Learning-Based Transactions Anomaly Prediction for Enhanced IoT Blockchain Network Security and Performance

KSII TRANSACTIONS ON INTERNET AND INFORMATION SYSTEMS 2024年第7期18卷 1986-2009页

作者： Abdullah, Nor Fadzilah Kairaldeen, Ammar Riadh Abu-Samah, Asma Nordin, Rosdiadee Univ Kebangsaan Malaysia Fac Engn & Built Environm Dept Elect Elect & Syst Engn Bangi 43600 Selangor Malaysia Wireless Res UKM Bangi 43600 Selangor Malaysia Sunway Univ Sch Engn & Technol Dept Engn Bandar Sunway 47500 Selangor Malaysia

The integration of blockchain technology with the rapid growth of Internet of Things (IoT) devices has enabled secure and decentralised data exchange. However, security vulnerabilities and performance limitations remain significant challenges in IoT blockchain networks. This work proposes a novel approach that combines transaction representation and machine learning techniques to address these challenges. Various clustering techniques, including kmeans, DBSCAN, Gaussian Mixture Models (GMM), and Hierarchical clustering, were employed to effectively group unlabelled transaction data based on their intrinsic characteristics. Anomaly transaction prediction models based on classifiers were then developed using the labelled data. Performance metrics such as accuracy, precision, recall, and F1-measure were used to identify the minority class representing specious transactions or security threats. The classifiers were also evaluated on their performance using balanced and unbalanced data. Compared to unbalanced data, balanced data resulted in an overall average improvement of approximately 15.85% in accuracy, 88.76% in precision, 60% in recall, and 74.36% in F1-score. This demonstrates the effectiveness of each classifier as a robust classifier with consistently better predictive performance across various evaluation metrics. Moreover, the k-means and GMM clustering techniques outperformed other techniques in identifying security threats, underscoring the importance of appropriate feature selection and clustering methods. The findings have practical implications for reinforcing security and efficiency in real-world IoT blockchain networks, paving the way for future investigations and advancements.

关键词： Anomaly prediction Blockchain clustering algorithms Fraud detection Internet of Things (IoT) Legitimate transactions Machine learning Privacy Security

来源：评论

学校读者我要写书评

暂无评论

Design of Hybrid Snake Optimizer Based Route Selection Approach for Unmanned Aerial Vehicles Communication

引用

IEEE ACCESS 2024年 12卷 54426-54434页

作者： Alkahtani, Hend Khalid Mahgoub, Hany Alotaibi, Faiz Abdullah Othman, Kamal M. Allafi, Randa Salama, Ahmed S. Princess Nourah bint Abdulrahman Univ Coll Comp & Informat Sci Dept Informat Syst POB 84428 Riyadh Saudi Arabia King Khalid Univ Coll Sci & Art Mahayil Dept Comp Sci Abha 62529 Saudi Arabia King Saud Univ Coll Arts Dept Informat Sci POB 28095 Riyadh Saudi Arabia Umm Al Qura Univ Dept Elect Engn Mecca 24382 Saudi Arabia Northern Border Univ Coll Sci & Arts Dept Comp & Informat Technol Ar Ar 73213 Saudi Arabia Future Univ Egypt Fac Engn & Technol Dept Elect Engn New Cairo 11845 Egypt

The effective usage of energy becomes crucial for the successful deployment and operation of unmanned aerial vehicles (UAVs) in different applications, such as surveillance, transportation, and communication networks. The increasing demand for UAVs in different industries such as agriculture, logistics, and emergency response has led to the development of more sophisticated and advanced UAVs. However, the limited onboard energy resource of UAVs poses a major problem for their long-term operation and endurance. In addition, artificial intelligence (AI) and machine learning (ML) could allow UAVs to make more informed and intelligent decisions regarding their operations, resulting in sustainable and more energy-efficient UAV deployment. This article designs a Hybrid Snake Optimizer-based Route Selection Approach for Unmanned Aerial Vehicles Communication (HSO-RSAUAVC) technique. The goal of the HSO-RSAUAVC technique is to explore and select optimal routes for UAV communication. In the presented HSO-RSAUAVC technique, the SO algorithm is integrated with Bernoulli Chaotic Mapping and Levy flight (LF) for enhanced performance. In addition, the HSO-RSAUAVC method derives a fitness function including residual energy (RE), distance, and UAV degree. By incorporating the HSO-RSAUAVC technique, we can dynamically adapt UAV paths to overcome obstacles, decrease communication interference, and optimize energy utilization. To validate the performance of the proposed model, a series of simulations were performed. The comparative result analysis illustrates the better performance of the HSO-RSAUAVC technique in improving the performance and reliability of UAV communication.

关键词： Autonomous aerial vehicles Heuristic algorithms Task analysis Planning clustering algorithms Adaptation models Statistics Routing Energy efficiency Metaheuristics Optimization methods Unmanned aerial vehicles routing snake optimizer energy efficiency fitness function

来源：评论

学校读者我要写书评

暂无评论

Enhancing Control Placement in SDN-IoT Networks Using the Louvain Algorithm and Betweenness-Centrality

引用

IEEE ACCESS 2024年 12卷 159775-159783页

作者： Abderrahmane, Amel Drid, Hamza Univ Batna 2 Dept Comp Sci LAMIE Lab Batna 05078 Algeria

The Internet of Things (IoT) continues to expand with an ever-growing number of devices. In this context, Software-Defined Networking (SDN) has emerged as a leading approach for effectively managing the dynamic and varied characteristics of IoT devices. Efficient controller placement is a crucial aspect of the SDN-IoT network, where the dynamic and diverse device connections demand responsive network management. This paper presents a novel approach for controller placement in SDN-IoT networks using the Louvain algorithm combined with the Betweenness-Centrality method. The proposed method leverages the Louvain algorithm to identify community structures within the network, representing groups of IoT devices with higher internal connectivity. Controllers are then strategically placed within these communities using the Betweenness-Centrality method. Our solution aims to minimize the number of controllers, reduce communication latency, and improve reliability. The modular nature of the Louvain algorithm allows for scalable and adaptive network segmentation, accommodating changes in network topology and IoT device dynamics. We conduct comprehensive simulations to evaluate the performance of our approach, comparing it with an optimal controller placement strategy. The results indicate enhanced network responsiveness, decreased latency, and improved overall efficiency. Our solution seamlessly adapts to network fluctuations, ensuring resilience in dynamic SDN-IoT environments.

关键词： Internet of Things Heuristic algorithms clustering algorithms Topology Software defined networking Scalability Partitioning algorithms Object recognition Delays Resource management Internet of Things (IoT) software-defined networking (SDN) SDN-IoT controller placement problem (CPP) Louvain algorithm betweenness-centrality

来源：评论

学校读者我要写书评

暂无评论

Improving Air Quality Zoning Through Deep Learning and Hyperlocal Measurements

引用

IEEE ACCESS 2024年 12卷 38700-38716页

作者： Fernandez, Eduardo Illueca Valera, Antonio Jesus Jara Breis, Jesualdo Tomas Fernandez Univ Murcia Dept Informat & Syst Murcia 30100 Spain Libelium LAB Res & Dev Dept Ceuti 30562 Spain

According to the Air Quality Directive 2008/50/EC, air quality zoning divides a territory into air quality zones where pollution and citizen exposure are similar and can be monitored using similar strategies. However, there is no standardized computational methodology to solve this problem, and only a few experiences in the Comunidad of Madrid based on chemistry transport models. In this study, we propose a methodological improvement based on the application of deep learning. Our method uses the CHIMERE-WRF air quality modelling system and adds a step that uses neural networks architectures to calibrate the simulations. We have validated our method in the Region of Murcia. The results obtained are promising given the values of the Pearson coefficient, obtaining r = 0.94 for NO2 and r = 0.95 for O-3, improving 86 % and 29 % the performances reported in the state of the art. In addition, the cluster score improves after applying neural networks, demonstrating that neural networks improve the consistency of clusters compared to the current air quality zoning. This opened new research opportunities based on the use of neural networks for dimension reduction in spatial clustering problems, and we were able to provide recommendations for a new measurement point in the Region of Murcia Air Quality Network.

关键词： Atmospheric modeling Air quality Computational modeling Neural networks Data models Biological system modeling Predictive models Artificial neural networks clustering methods Deep learning artificial neural networks atmospheric modeling clustering algorithms deep learning

来源：评论

学校读者我要写书评

暂无评论

Efficient Detection of k-Plex Structures in Large Graphs Through Constraint Learning

引用

IEEE TRANSACTIONS ON COMPUTATIONAL SOCIAL SYSTEMS 2024年

作者： Hung, Hui-Ju Lu, Chia-Hsun Huang, Yun-Ya Chang, Ming-Yi Ho, Ya-Chi Shen, Chih-Ya Natl Cent Univ Dept Comp Sci & Informat Engn Taoyuan 320317 Taiwan Natl Tsing Hua Univ Dept Comp Sci Hsinchu 300044 Taiwan Fu Jen Catholic Univ Dept Sociol Taipei 242062 Taiwan Ind Technol Res Inst ITRI Hsinchu 310401 Taiwan

The k-plex is a popular definition of communities in networks, offering more flexibility than cliques by allowing each node to miss up to k connections. However, finding k-plexes in large graphs is a theoretically challenging task due to the large number of possible k-plexes. In this article, we propose a novel approach for detecting k-plexes under various sizes and time constraints using an automated strategy to learn bounds, called the constraint learning and bounding (CLB) method. Specifically, our proposed CLB approach, leverages the concept of constraint learning to develop a mixed integer linear programming (MILP) instance as a model to learn a bounding strategy in the branch-and-bound process. The variables in the MILP instances correspond to the natural properties of the k-plex problem. Unlike previous works, we focus on learning the bounding strategy rather than learning the branching strategy. Thus, the strategy learned by our proposed approach avoids visiting infeasible solutions, which accelerates the branch-and-bound algorithm and reduces the computational load. To evaluate our approach, we conduct experiments on various real graphs to validate the superiority and the generality of our proposed approach. In summary, our approach offers an effective and efficient solution for detecting k-plexes under various conditions.

关键词： Optimization Computational modeling Computational efficiency Training data Time factors Social networking (online) Manuals Complexity theory clustering algorithms Unsupervised learning Branch-and-bound constraint learning k-plex mixed integer linear programming (MILP)

来源：评论

学校读者我要写书评

暂无评论

Comprehensive Analysis of clustering Methods: Focusing on Fuzzy clustering

Comprehensive Analysis of Clustering Methods: Focusing on Fu...

引用

2025 International Conference on Multi-Agent Systems for Collaborative Intelligence, ICMSCI 2025

作者： Salve, Vaishali P. Ghatule, Magan P. Sinhgad College Of Science Dept. of Electronic Science Pune India

ISBN: (纸本)9798331509828

The objective of clustering algorithms is to group or cluster data according to the underlying patterns and commonalities in the data. They are crucial to many aspects of modern life, including social media, healthcare, marketing and e-commerce, and data organization and analysis. There are many clustering methods available, and new ones are always being developed. There are strengths and limitations associated with each algorithm, and there is currently no one method that works for every task. In this study, the researchers analyzed existing clustering algorithms and categorized popular algorithms along five dimensions: fundamental various clustering techniques, fuzzy clustering, the proposed Fuzzy C-Mean (FCM) clustering, algorithms are classified based on fundamental clustering method and data size, and it also offers a comparison of studies comparing Fuzzy C-Means (FCM) clustering with other traditional clustering algorithms. This classification makes it easier for researchers to understand clustering algorithms from different angles and helps in the identification of algorithms that are appropriate for resolving particular problems. Lastly, they discussed the current condition of clustering algorithms now and possible future developments. The researchers additionally discussed and noted the field's open difficulties and unresolved issues. © 2025 IEEE.

关键词： clustering algorithms

来源：评论

学校读者我要写书评

暂无评论

A Game-Theoretic Approach for User-Centric AP Selection in Cell-Free Massive MIMO Under Dynamic User Demand Environment

引用

IEEE WIRELESS COMMUNICATIONS LETTERS 2025年第3期14卷 876-880页

作者： Wei, Chen Tian, Hui Xu, Kui Xia, Xiaochen Xie, Wei Li, Chunguo Army Engn Univ PLA Coll Commun Engn Nanjing 210007 Peoples R China Army Engn Univ PLA Coll Field Engn Nanjing 210007 Peoples R China Southeast Univ Sch Informat Sci & Engn Nanjing 210096 Peoples R China

This letter investigates user-centric (UC) access point (AP) selection for sum spectral efficiency (SE) maximization in cell-free massive multiple-input multiple-output (MIMO) systems under dynamic user demand environment. The AP selection problem can be modeled as a stochastic game with local altruistic characteristics. It can be proved to be an exact potential game (EPG), which further clarifies the existence of pure strategy Nash equilibrium (NE). Then, we propose a stochastic learning-based distributed UC AP selection algorithm (DUAPSA) to achieve the NE. Finally, simulation results are conducted to verify the validity of the DUAPSA algorithm in dynamic user demand environment.

关键词： Stochastic processes Games Heuristic algorithms Massive MIMO Interference Probability Vectors Optimization clustering algorithms Central Processing Unit AP selection cell-free massive MIMO dynamic user demand environment sum SE maximization stochastic game stochastic learning

来源：评论

学校读者我要写书评

暂无评论

SIG: Graph-Based Cancer Subtype Stratification With Gene Mutation Structural Information

引用

IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS 2024年第6期21卷 1752-1764页

作者： Zhang, Chengcheng Li, Wei Deng, Ming Jiang, Yizhang Cui, Xiaohui Chen, Ping Jiangnan Univ Sch Artificial Intelligence & Comp Sci Wuxi 214126 Jiangsu Peoples R China Jiangnan Univ Jiangsu Key Lab Media Design & Software Technol Wuxi 214126 Jiangsu Peoples R China Jiangnan Univ Engn Res Ctr Intelligent Technol Healthcare Minist Educ Wuxi 214126 Jiangsu Peoples R China Wuhan Univ Zhongnan Hosp Dept Radiol Wuhan 430071 Peoples R China Wuhan Univ Sch Cyber Sci & Engn Wuhan 430071 Peoples R China Univ Massachusetts Boston Dept Engn Boston MA 02125 USA

Somatic tumors have a high-dimensional, sparse, and small sample size nature, making cancer subtype stratification based on somatic genomic data a challenge. Current methods for improving cancer clustering performance focus on dimension reduction, integrating multi-omics data, or generating realistic samples, yet ignore the associations between mutated genes within the patient-gene matrix. We refer to these associations as gene mutation structural information, which implicitly includes cancer subtype information and can enhance subtype clustering. We introduce a novel method for cancer subtype clustering called SIG(Structural Information within Graph). As cancer is driven by a combination of genes, we establish associations between mutated genes within the same patient sample, pair by pair, and use a graph to represent them. An association between two mutated genes corresponds to an edge in the graph. We then merge these associations among all mutated genes to obtain a structural information graph, which enriches the gene network and improves its relevance to cancer clustering. We integrate the somatic tumor genome with the enriched gene network and propagate it to cluster patients with mutations in similar network regions. Our method achieves superior clustering performance compared to SOTA methods, as demonstrated by clustering experiments on ovarian and LUAD datasets.

关键词： Cancer Tumors Genomics Bioinformatics clustering algorithms Germanium Task analysis Cancer clustering feature selection gene mutat ion structural information genomic data graph

来源：评论

学校读者我要写书评

暂无评论

A BCI System for Imagined Speech Classification Based on Optimization Theory

引用

IEEE TRANSACTIONS ON CONSUMER ELECTRONICS 2024年第4期70卷 6679-6690页

作者： Zheng, Xiao-Ben Ling, Bingo Wing-Kuen Guangdong Univ Technol Sch Informat Engn Guangzho 510006 Peoples R China

Electroencephalograms (EEGs) are used for establishing a connection between the human brain and the outside environment, so they are widely used in the brain computer interface (BCI). Nowadays, the imagined speech (IS) is a highly promising paradigm of the BCI. It can be used for controlling the external devices directly. However, the features for performing the IS are unknown. Hence, the numerous features are extracted. As a result, the dimension of the feature vectors is extremely large. To reduce the required computation, the clustering is required to be performed in the low dimensional space. Under this circumstance, the transform matrix affects both the dimensional reduction part and the clustering part. In fact, finding the transform matrix and the clustering centers under this scenario is challenging. To tackle this difficulty, this paper provides a modified joint principal component analysis (PCA) and k means algorithm for performing the IS. Here, the interclass separation among the feature vectors is also taken into an account of the problem formulation. In particular, the problem is formulated as a nonconvex constrained optimization problem. The total two norm reconstruction error of the feature vectors as well as the total two norm differences between the feature vectors and the clustering centers in the low dimensional space and the total two norm differences among the clustering centers are minimized subject to the orthogonality of the transform matrix. The numerical computer simulations are conducted based on the multi-class IS classification database. The obtained results show that our proposed method outperforms the various states of the art methods in terms of the clustering accuracy and the average required execution time. Overall, using the BCI system for performing the imagined speech classification plays an important role in the consumer electronics area.

关键词： clustering algorithms Optimization Consumer electronics Transforms Principal component analysis Machine learning algorithms Classification algorithms Vectors Electroencephalography Dimensionality reduction Semi-supervised imagined speech k means clustering in the low dimensional space principal component analysis linear discriminant analysis nonconvex constrained optimization problem

来源：评论

学校读者我要写书评

暂无评论

Improved Application Placement in Fog Environments Through Parallel Collaboration

引用

IEEE ACCESS 2024年 12卷 177379-177397页

作者： Aljohani, Aisha Sakellariou, Rizos Yanbu Ind Coll Dept Comp Sci & Engn Yanbu 46454 Saudi Arabia Univ Manchester Dept Comp Sci Manchester M13 9PL Lancs England

The real-time Quality of Service (QoS) constraints of Internet-of-Things (IoT) applications have motivated the development of distributed systems environments such as fog computing. A challenging issue in fog computing is to identify placement plans by which the IoT applications are placed on fog devices for execution, a problem known as the application placement problem. With a potentially huge number of fog devices and IoT applications, it makes sense to solve this problem in a decentralized manner, where an independent placement problem is solved in parallel for different clusters of fog devices. To cope with the highly dynamic nature of the IoT networks, clusters lacking a sufficient number of fog devices to accommodate all the submitted IoT applications can propagate undeployed applications to other clusters for execution, potentially leading to uncertain fulfillment of QoS constraints. The challenge becomes how to enhance the probability of placing the propagated IoT applications within their predefined QoS constraints, i.e., response time delays. This paper proposes an application placement approach, called PColl, that aims to minimize the total number of delay violations for applications, specifically for the propagated applications, through parallel collaboration. The proposed approach utilizes a heuristic-based algorithm that enhances the probability of placing the propagated applications within their delays through parallel collaboration. This is achieved by duplicating the placement requests of these applications, which enables parallel searching for fog devices. Additionally, this paper proposes a collaboration algorithm where various architecture entities collaborate to manage the problem of parallel searching. The experimental results demonstrate the effectiveness of the proposed approach in minimizing the percentage of applications that experience delay violations, as compared to other placement approaches.

关键词： Internet of Things Delays Computer architecture Quality of service Collaboration Databases Computational modeling clustering algorithms Reviews Resource management Application/service placement fog clusters fog computing propagated applications delay-sensitive applications iFogsim quality of service

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：