检索结果-内蒙古大学图书馆

Privacy-Aware Federated Learning Framework for IoT Security Using Chameleon Swarm Optimization and Self-Attentive Variational Autoencoder

引用

computer Modeling in Engineering & sciences 2025年第4期143卷 849-873页

作者： Saad Alahmari Abdulwhab Alkharashi Department of Computer Science Applied CollegeNorthern Border UniversityArar91431Saudi Arabia Department of Computer Science College of Computing and InformaticsSaudi Electronic UniversityRiyadh11673Saudi Arabia

The Internet of Things(IoT)is emerging as an innovative phenomenon concerned with the development of numerous vital *** the development of IoT devices,huge amounts of information,including users’private data,are *** systems face major security and data privacy challenges owing to their integral features such as scalability,resource constraints,and *** challenges are intensified by the fact that IoT technology frequently gathers and conveys complex data,creating an attractive opportunity for *** address these challenges,artificial intelligence(AI)techniques,such as machine learning(ML)and deep learning(DL),are utilized to build an intrusion detection system(IDS)that helps to secure IoT *** learning(FL)is a decentralized technique that can help to improve information privacy and performance by training the IDS on discrete linked *** delivers an effectual tool to defend user confidentiality,mainly in the field of IoT,where IoT devices often obtain privacy-sensitive personal *** study develops a Privacy-Enhanced Federated Learning for Intrusion Detection using the Chameleon Swarm Algorithm and Artificial Intelligence(PEFLID-CSAAI)*** main aim of the PEFLID-CSAAI method is to recognize the existence of attack behavior in IoT ***,the PEFLIDCSAAI technique involves data preprocessing using Z-score normalization to transformthe input data into a beneficial ***,the PEFLID-CSAAI method uses the Osprey Optimization Algorithm(OOA)for the feature selection(FS)*** the classification of intrusion detection attacks,the Self-Attentive Variational Autoencoder(SA-VAE)technique can be ***,the Chameleon Swarm Algorithm(CSA)is applied for the hyperparameter finetuning process that is involved in the SA-VAE model.A wide range of experiments were conducted to validate the execution of the PEFLID-CSAAI *** simulated outcomes demonstrated that the PEFLID-CSAAI

关键词： Federated learning internet of things artificial intelligence chameleon swarm algorithm intrusion detection system healthcare IoT devices

来源：评论

学校读者我要写书评

暂无评论

Contemporary Study for Detection of COVID-19 Using Machine Learning with Explainable AI

引用

computers, Materials & Continua 2024年第7期80卷 1075-1104页

作者： Saad Akbar Humera Azam Sulaiman Sulmi Almutairi Omar Alqahtani Habib Shah Aliya Aleryani Department of Computer Science University of KarachiKarachi75270Pakistan Department of Computing Faculty of Engineering Science and TechnologyHamdard UniversityKarachi75540Pakistan Department of Health Informatics College of Applied Medical SciencesQassim UniversityQassim51452Saudi Arabia Department of Computer Science College of Computer ScienceKing Khalid UniversityAbha61421Saudi Arabia

The prompt spread of COVID-19 has emphasized the necessity for effective and precise diagnostic *** this article,a hybrid approach in terms of datasets as well as the methodology by utilizing a previously unexplored dataset obtained from a private hospital for detecting COVID-19,pneumonia,and normal conditions in chest X-ray images(CXIs)is proposed coupled with Explainable Artificial Intelligence(XAI).Our study leverages less preprocessing with pre-trained cutting-edge models like InceptionV3,VGG16,and VGG19 that excel in the task of feature *** methodology is further enhanced by the inclusion of the t-SNE(t-Distributed Stochastic Neighbor Embedding)technique for visualizing the extracted image features and Contrast Limited Adaptive Histogram Equalization(CLAHE)to improve images before extraction of ***,an AttentionMechanism is utilized,which helps clarify how the modelmakes decisions,which builds trust in artificial intelligence(AI)*** evaluate the effectiveness of the proposed approach,both benchmark datasets and a private dataset obtained with permissions from Jinnah PostgraduateMedical Center(JPMC)in Karachi,Pakistan,are *** 12 experiments,VGG19 showcased remarkable performance in the hybrid dataset approach,achieving 100%accuracy in COVID-19 *** classification and 97%in distinguishing normal ***,across all classes,the approach achieved 98%accuracy,demonstrating its efficiency in detecting COVID-19 and differentiating it fromother chest disorders(Pneumonia and healthy)while also providing insights into the decision-making process of the models.

关键词： COVID-19 detection deep neural networks support vector machine CXIs InceptionV3 VGG16 VGG19 t-SNE embedding CLAHE attention mechanism XAI

来源：评论

学校读者我要写书评

暂无评论

Addressing Imbalance in Health Datasets: A New Method NR-Clustering SMOTE and Distance Metric Modification

引用

computers, Materials & Continua 2025年第2期82卷 2931-2949页

作者： Hairani Hairani Triyanna Widiyaningtyas Didik Dwi Prasetya Afrig Aminuddin Department of Electrical Engineering and Informatics Faculty of EngineeringUniversitas Negeri MalangMalang65145Indonesia Department of Computer Science Universitas BumigoraMataram83127Indonesia Department of Computer Graphic and Multimedia Faculty of ComputingCollege of Computing and Applied SciencesUniversiti Malaysia Pahang Al-Sultan AbdullahPekan26600Malaysia

An imbalanced dataset often challenges machine learning, particularly classification methods. Underrepresented minority classes can result in biased and inaccurate models. The Synthetic Minority Over-Sampling Technique (SMOTE) was developed to address the problem of imbalanced data. Over time, several weaknesses of the SMOTE method have been identified in generating synthetic minority class data, such as overlapping, noise, and small disjuncts. However, these studies generally focus on only one of SMOTE’s weaknesses: noise or overlapping. Therefore, this study addresses both issues simultaneously by tackling noise and overlapping in SMOTE-generated data. This study proposes a combined approach of filtering, clustering, and distance modification to reduce noise and overlapping produced by SMOTE. Filtering removes minority class data (noise) located in majority class regions, with the k-nn method applied for filtering. The use of Noise Reduction (NR), which removes data that is considered noise before applying SMOTE, has a positive impact in overcoming data imbalance. Clustering establishes decision boundaries by partitioning data into clusters, allowing SMOTE with modified distance metrics to generate minority class data within each cluster. This SMOTE clustering and distance modification approach aims to minimize overlap in synthetic minority data that could introduce noise. The proposed method is called “NR-Clustering SMOTE,” which has several stages in balancing data: (1) filtering by removing minority classes close to majority classes (data noise) using the k-nn method;(2) clustering data using K-means aims to establish decision boundaries by partitioning data into several clusters;(3) applying SMOTE oversampling with Manhattan distance within each cluster. Test results indicate that the proposed NR-Clustering SMOTE method achieves the best performance across all evaluation metrics for classification methods such as Random Forest, SVM, and Naїve Bayes, compared t

关键词： SMOTE modification Clustering-SMOTE manhattan distance

来源：评论

学校读者我要写书评

暂无评论

An efficient feature selection and classification system for microarray cancer data using genetic algorithm and deep belief networks

引用

Multimedia Tools and Applications 2025年第8期84卷 4393-4434页

作者： Lawrence, Morolake Oladayo Jimoh, Rasheed Gbenga Yahya, Waheed Babatunde Department of Computer Science Faculty of Computing & Applied Sciences Baze University Abuja Nigeria Department of Computer Science Faculty of Communication & Information Sciences University of Ilorin Kwara State Ilorin Nigeria Department of Statistics Faculty of Physical Sciences University of Ilorin Kwara State Ilorin Nigeria

Cancer is one of the most devastating health conditions in the world. In the diagnosis and treatment of the various forms of cancer illness, studies have shown that early detection of the cancer by clinical methods usually takes a considerable lengthy time. This informs searching for an alternative non-clinical diagnosis of cancer cases using microarray technology. Therefore, this study develops an efficient feature selection and classification method for high-dimensional microarray cancer data by combining Genetic Algorithms (GA) and Deep Belief Networks (DBN). The study employed a GA for selecting the most informative gene biomarker and DBN for the classification of biological samples. In a Monte Carlo experiment, the simulated and real-life microarray datasets were partitioned into 95% training and 5% test samples at 100 to 1000 epochs. The classifier was constructed using the training datasets while its efficiency was assessed on the test sample using the Misclassification Error Rate, Sensitivity, Specificity, and Receiver Operating Characteristic Analysis. The GA and DBN were implemented using Caret and Deepnet (R statistical packages). The proposed GADBN method was implemented using simulated and real-life datasets, which yielded out-of-bag average classification accuracy of 98.8% and 93.1% respectively. The proposed GADBN outperformed the other existing classifiers;and the GADBN method was more efficient with a smaller Misclassification Error Rate (MER) for Leukaemia 1 (0.18), Prostate 1 (0.35) and Prostate 3 (0.09) datasets than some of the existing methods under various performance indices considered. The proposed model is a powerful and effective instrument for identifying useful features in microarray cancer data and classifying the cancer types accordingly. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

A modified grey wolf optimizer with multi-solution crossover integration algorithm for feature selection

引用

Journal of Ambient Intelligence and Humanized computing 2025年第1期16卷 329-345页

作者： Ihsan, Muhammad Din, Fakhrud Zamli, Kamal Z. Ghadi, Yazeed Yasin Alahmadi, Tahani Jaser Innab, Nisreen Faculty of IT Department of Computer Science and IT University of Malakand Khyber Pakhtunkhwa Chakdara18800 Pakistan Faculty of Computing College of Computing and Applied Sciences Universiti Malaysia Pahang Pahang Pekan26600 Malaysia Department of computer science and software engineering Al Ain University Abu Dhabi12555 United Arab Emirates Department of Information SystemsCollege of Computer and Information Sciences Princess Nourah bint Abdulrahman University Riyadh84428 Saudi Arabia Department of Computer Science and Information SystemsCollege of Applied Sciences Al-Maarefa University Diriyah Riyadh13713 Saudi Arabia

Feature selection helps eradicate redundant features which is essential to mitigate the curse of dimensionality when a machine-learning model deals with high-dimensional datasets. Grey Wolf Optimizer (GWO) is a swarm-based algorithm that simulates the wolves’ hunting behavior. Although very efficient, GWO faces some limitations which may cause premature convergence and/or local optima trapping. Moreover, GWO relies mainly on the three best wolves, limiting its potential for diverse exploration and exploitation. This work proposes an improved version of GWO namely, a modified grey wolf optimizer with multi-solution crossover integration (MGWO-MCI) algorithm. MGWO-MCI algorithm incorporates a multi-solution strategy that evolves new potential solutions in the optimization process. A crossover operation is performed between the new wolves and the existing hierarchy, reforming the position-updating process. MGWO-MCI utilizes this position-updating process using two different approaches. The first approach named MGWO-MCI-I expands the additional wolves’ role to both exploration and exploitation whereas the second approach named MGWO-MCI-II incorporates their role to exploration only. These approaches are evaluated and tested using 18 datasets and an Intrusion detection dataset NSL-KDD for feature selection. Statistically, the results are analyzed through the Wilcoxon test, which shows the superiority of MGWO-MCI-II. MGWO-MCI-II outperforms others with an accuracy of 98.6% on NSL-KDD and achieves 55.5% overall best outcomes on other datasets. Moreover, the MGWO-MCI was evaluated on two constrained optimization problems, the pressure vessel and welded beam design validating its effectiveness and adaptability in solving different optimization problems. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2025.

关键词： Constrained optimization

来源：评论

学校读者我要写书评

暂无评论

An EfficientNet to Classify Monkeypox-Comparable Skin Lesions Using Transfer Learning

An EfficientNet to Classify Monkeypox-Comparable Skin Lesion...

引用

2023 IEEE International Conference on Omni-Layer Intelligent Systems, COINS 2023

作者： Hossen, Md. Rifat Alfaz, Nazia Sami, Adnan Tanim, Sharia Arfin Bin Sarwar, Talha Islam, Md. Kamrul Department of Computer Science and Engineering Dhaka Bangladesh Department of Computer Science Dhaka Bangladesh College of Computing and Applied Science Universiti Malaysia Pahang Faculty of Computing Pekan Malaysia

ISBN: (纸本)9798350346473

Monkeypox is an infectious illness caused by the DNA-based monkeypox virus, which has raised public health concerns due to its rapid transmission to over 50 countries. Direct physical interaction with infected humans or infected animals is the main reason behind the spread of this virus. The appearance of skin problems such as smallpox and rashes are the most frequently reported symptoms of this virus. Since cases of monkeypox are increasing rapidly around the world, stopping the spread of this zoonosis by providing early diagnosis and treatment is crucial before the emergence of a pandemic similar to COVID-19. In this study, we aim to propose a transfer learning-based approach using the EfficientNet-B0 architecture to identify monkeypox subjects by using skin lesion image samples. However, distinguishing monkeypox from other comparable infectious skin illnesses like chickenpox and measles is challenging. Therefore, additionally, this study identifies other diseases that also cause blisters and rashes on the skin, like chickenpox, and measles. During the data distribution phase, 5-fold cross-validation is used to validate the work's reliability by assuring that every sample is utilized for training and validation. For the evaluation of the model's classification performance, accuracy and loss are recorded for each training epoch. Moreover, precision, recall, F1-score, and confusion matrix are generated upon completion of the model training. This proposed approach is experimented on a public dataset and has shown remarkable performance by providing an overall 96.53% classification accuracy, 96.57% precision, 96.53% recall, and 96.52% F1-score. © 2023 IEEE.

关键词： Viruses

来源：评论

学校读者我要写书评

暂无评论

An Adaptive on Demand Modified Ant Colony Optimization Routing for VANETs 23th

An Adaptive on Demand Modified Ant Colony Optimization Routi...

引用

23rd International Conference on Intelligent Systems Design and Applications, ISDA 2023

作者： Kumari, J. Jesy Janet Thangam, S. Raja, A. Saleem Department of Computer Science and Engineering Amrita School of Computing Amrita Vishwa Vidyapeetham Bengaluru India Department of Computer Science and Engineering University of Technology and Applied Science-Shinas Shinas Oman

ISBN: (纸本)9783031648496

Around the world, vehicle accidents claim many priceless lives. To tackle this problem, the Vehicular ad hoc networks (VANETs) are the most effective solution. Effective communication between vehicular nodes in VANETs is extremely important because of the rapid mobility of vehicles and different network topology. Currently, VANET mostly serves the public security, stability, information for voyageurs, and traffic Management, control of congestion, guidance, etc. Chaos on roadways can occur at any moment, anyplace. It is a situation in which the flow of traffic regulations might cease to function. Collisions, maintaining roads, violation of legal rules, etc., so to evade the congestion in the surrounding region. VANETs represent a dynamic and challenging communication environment characterized by high mobility, intermittent connectivity, and rapid changes in network topology. Efficient and reliable routing is crucial for enabling diverse applications such as traffic management, safety communication, and infotainment services in VANETs. In this research, propose a multi-faceted approach to improve routing performance in VANETs. An adaptive routing algorithm that considers vehicle mobility patterns and dynamically adjusts communication routes based on the anticipated movement of neighboring vehicles such as Modified Ant Colony Optimization (MACO) technique. This adaptability enhances the robustness of the network against frequent topology changes and improves the overall packet delivery ratio, end-to-end delay, and network reliability compared to traditional routing protocols. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

关键词： Vehicular ad hoc networks

来源：评论

学校读者我要写书评

暂无评论

Ensuring Zero Trust IoT Data Privacy: Differential Privacy in Blockchain using Federated Learning

引用

IEEE Transactions on Consumer Electronics 2024年第1期71卷 1-1页

作者： Hussain, Altaf Akbar, Wajahat Hussain, Tariq Bashir, Ali Kashif Dabel, Maryam M. Al Ali, Farman Yang, Bailin School of Electronic and Control Engineering Chang’an University Xi’an China School of Computer Science and Technology and School of Mathematics and Statistics Zhejiang Gongshang University Hangzhou China Department of Computing and Mathematics Manchester Metropolitan University UK Department of Computer Science and Engineering College of Computer Science and Engineering University of Hafr Al Batin Saudi Arabia Department of Applied AI School of Convergence College of Computing and Informatics Sungkyunkwan University Seoul South Korea

In the increasingly digitized world, the privacy and security of sensitive data shared via IoT devices are paramount. Traditional privacy-preserving methods like k-anonymity and ldiversity are becoming outdated due to technological advancements. In addition, data owners often worry about misuse and unauthorized access to their personal information. To address this, we propose a secure data-sharing framework that uses local differential privacy (LDP) within a permissioned blockchain, enhanced by federated learning (FL) in a zero-trust environment. To further protect sensitive data shared by IoT devices, we use the Interplanetary File System (IPFS) and cryptographic hash functions to create unique digital fingerprints for files. We mainly evaluate our system based on latency, throughput, privacy accuracy, and transaction efficiency, comparing the performance to a benchmark model. The experimental results show that the proposed system outperforms its counterpart in terms of latency, throughput, and transaction efficiency. The proposed model achieved a lower average latency of 4.0 seconds compared to the benchmark model’s 5.3 seconds. In terms of throughput, the proposed model achieved a higher throughput of 10.53 TPS (transactions per second) compared to the benchmark model’s 8 TPS. Furthermore, the proposed system achieves 85% accuracy, whereas the counterpart achieves only 49%. IEEE

关键词： Anonymity

来源：评论

学校读者我要写书评

暂无评论

A Novel Congestion Control Scheme Using Fuzzy Logic Systems to Enhance the Path Selection Criteria in Routing Protocols for Low-Power and Lossy Networks on the Internet of Things

引用

SN computer science 2024年第5期5卷 1-31页

作者： Aimtongkham, Phet Musikawan, Pakarat Kongsorot, Yanika So-In, Chakchai Applied Network Technology (ANT) Department of Computer Science College of Computing Khon Kaen University Khon Kaen 40002 Thailand Advanced Intelligent Interdisciplinary Integration (AIII) Department of Computer Science College of Computing Khon Kaen University Khon Kaen 40002 Thailand

In routing protocols for low-power and lossy (RPL)-based Internet of Things (IoT) networks, congestion control is essential for ensuring efficient and dependable communications with energy awareness. As the number of devices and the amount of data traffic increase, congestion typically occurs, resulting in degraded network performance, i.e., increased packet losses and delays with decreased energy efficiency. Many parameters relating to congestion behaviors have been considered and dynamically changed, as have the possible transmission paths either within a single destination-oriented directed acyclic graph (DoDAG) or across different DoDAGs. Thus, this research investigates congestion mitigation methods for RPL-IoT networks based on a fuzzy logic system (FLS) by proposing a novel FLS scheme to determine the selection criteria for conducting path management in three FLS components: (1) inside/outside the DoDAG selection process, (2) inside the path selection process within a single DoDAG (the so-called inside DoDAG) for high or to-be-high loads, and (3) inside the path selection process across the DoDAGs (the so-called outside DoDAG) during severe congestion applying the concept of a multi-DoDAG node (MDN) to determine both a suitable MDN and data traffic for alternate path selection. Simulation results demonstrate that the proposed method effectively controls congestion levels, decreases packet losses and latency, and attains improved energy efficiency in comparison with the state-of-the-art congestion control schemes in RPL-IoT networks. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2024.

关键词： Congestion controls Fuzzy logic systems Internet of Things Objective function Path selection Routing protocol for low-power and lossy networks

来源：评论

学校读者我要写书评

暂无评论

Enhancing Gastrointestinal Image Classification: A Fusion of CNN and Vision Trasformers 2

Enhancing Gastrointestinal Image Classification: A Fusion of...

引用

2nd IEEE International Conference on Contemporary computing and Communications, InC4 2024

作者： Ganesh, Bachu Annamalai, R. Jayapal, Senthil Amrita School of Computing Amrita Vishwa Vidyapeetham Department of Computer Science and Engineering Chennai India Collage of Computing and Information Science University of Technology and Applied Sciences Ibra Oman

ISBN: (纸本)9798350383652

In order to meet the pressing demand for early diagnosis in healthcare, we proposed a unique hybrid architecture in this study that is intended for the categorization of gastrointestinal (GI) illnesses. With fewer trainable parameters, faster training times, and less storage needed for model weights, our suggested architecture prioritizes efficiency achieves an impressive 94% accuracy. To effectively diagnose GI disorders, this design combines convolutional neural networks (CNNs) with Transformers. The creation of this hybrid architecture, which boasts excellent accuracy in illness categorization, decreased storage requirements, and efficiency in training, is one of the research's accomplishments. Our proposed model demonstrates its flexibility for many uses, encouraging simple implementation and expandability. Extensive tests performed on a large dataset of endoscopic images highlight our hybrid architecture's better performance over current techniques. With this study, diagnostics for gastrointestinal diseases have advanced significantly, opening the door to future diagnostic systems that will be more precise and effective. The proposed architecture has outperformed the traditional pre-trained models both in classification accuracy and time to train the model with no weights. © 2024 IEEE.

关键词： Endoscopy

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

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

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：