检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Wang, Xiaoyang Yang, Christopher C. College of Computing and Informatics Drexel University Philadelphia United States

Artificial intelligence (AI) systems in healthcare have demonstrated remarkable potential to improve patient outcomes. However, if not designed with fairness in mind, they also carry the risks of perpetuating or exacerbating existing health disparities. Although numerous fairness-enhancing techniques have been proposed, most focus on a single sensitive attribute and neglect the broader impact that optimizing fairness for one attribute may have on the fairness of other sensitive attributes. In this work, we introduce a novel approach to multi-attribute fairness optimization in healthcare AI, tackling fairness concerns across multiple demographic attributes concurrently. Our method follows a two-phase approach: initially optimizing for predictive performance, followed by fine-tuning to achieve fairness across multiple sensitive attributes. We develop our proposed method using two strategies, sequential and simultaneous. Our results show a significant reduction in Equalized Odds Disparity (EOD) for multiple attributes, while maintaining high predictive accuracy. Notably, we demonstrate that single-attribute fairness methods can inadvertently increase disparities in non-targeted attributes whereas simultaneous multi-attribute optimization achieves more balanced fairness improvements across all attributes. These findings highlight the importance of comprehensive fairness strategies in healthcare AI and offer promising directions for future research in this critical area. © 2025, CC BY.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

IAPS: Decreasing Software-Based Packet Steering Overhead Through Interrupt Reduction

IAPS: Decreasing Software-Based Packet Steering Overhead Thr...

引用

International Conference on Information Networking

作者： Maike Helbig Younghoon Kim College of Computing and Informatics Sungkyunkwan University Suwon South Korea

ISBN: (数字)9798331506940

ISBN: (纸本)9798331506957

Packet Steering is often performed in hardware, rendering the software-based packet steering mechanisms of the Linux kernel obsolete. However, the ability to redistribute packets later during network processing could help achieve a more parallelized network stack. Unfortunately, existing software-based packet steering schemes yield minimal performance gain at high communication costs. This paper proposes Interrupt Avoidance Packet Steering (IAPS), a novel packet steering scheme that reduces software-based packet steering overhead. IAPS decreases communication costs by avoiding hardware interrupt triggers during packet steering. IAPS increases application throughput by up to $\mathbf{4. 3 \%}$ and the packet per hardware interrupt ratio by up to 4 x.

关键词： Costs Linux Performance gain Throughput Rendering (computer graphics) Hardware Kernel

来源：评论

学校读者我要写书评

暂无评论

Decision-Aid Framework for Face Authentication Detection Using ResNext50 and BiLSTM to Enhance Media Integrity

引用

IEEE Access 2025年 13卷 89858-89873页

作者： Alrawahneh, Ayat Abd-Muti Abdullah, Siti Norul Huda Sheikh Abuain, Tarik Abdullah, Sharifah Nurul Asyikin Syed Taylor, Sarah Khadijah Suhaimi, Nur Hanis Sabrina Universiti Kebangsaan Malaysia Faculty of Information Science and Technology Bangi 43600 Malaysia Saudi Electronic University College of Computing and Informatics Riyadh 11673 Saudi Arabia CyberSecurity Malaysia Cyberjaya 63000 Malaysia

This paper presents a decision-aid framework for face authentication detection that integrates ResNext50 with Bidirectional Long Short-Term Memory (BiLSTM) networks to enhance media integrity and improve deepfake detection. Unlike conventional approaches that focus solely on spatial features, the proposed hybrid model incorporates temporal analysis, enabling the detection of subtle manipulations distributed across sequential video frames. ResNext50 serves as a robust spatial feature extractor from cropped facial regions, while BiLSTM captures bidirectional temporal dependencies, enriching the model's contextual understanding and temporal sensitivity. The model was evaluated on three benchmark datasets - FaceForensics++, DFDC, and Celeb-DF - and demonstrated superior performance in classifying real versus manipulated facial content. The hybrid ResNext50 + BiLSTM model achieved a final accuracy of 96.11%, precision of 96.42%, recall of 96.68%, F1-score of 96.23%, and an AUC of 98.89%, outperforming the standalone ResNext50 and other state-of-the-art CNN-based approaches. These improvements reflect a notable reduction in both false positives and false negatives, which is critical for real-world face authentication systems. The proposed framework offers a scalable and robust solution for applications in digital forensics, media verification, and cybersecurity, where maintaining trust in visual content is paramount. This study underscores the pivotal role of temporal modeling in improving the reliability of face manipulation detection. Future work will explore advanced fusion strategies, lightweight architectures for real-time deployment, and cross-dataset generalization to enhance adaptability in diverse, real-world scenarios. © 2013 IEEE.

关键词： BiLSTM Decision-Aid Face Authentication Detection Media Integrity ResNext50

来源：评论

学校读者我要写书评

暂无评论

Edge AI-powered marine pollution classification with customized CNN model

引用

Neural computing and Applications 2025年第9期37卷 6449-6463页

作者： Palanisamy, Sanjai Bonny, Talal Nasir, Nida Al Shabi, Mohammad Al Shammaa, Ahmed University of Sharjah Sharjah United Arab Emirates Khorfakkan University Sharjah United Arab Emirates College of Computing and Informatics University of Sharjah Sharjah United Arab Emirates Mechanical Department College of Engineering University of Sharjah Sharjah United Arab Emirates Department of Electronics and Communication Engineering Karunya Institute of Technology and Sciences Coimbatore India

The increasing production of disposable plastic products contributes greatly to marine pollution and its impact on the marine ecosystem and organisms consuming ocean-derived food. To address this issue, this paper proposes a new customized convolutional neural network (CNN) model for categorizing the level of marine pollution in underwater ocean regions using image classification. The customized CNN model is developed and compared with five preexisting models, including DenseNet121, Inception-ResNetV2, InceptionV3, VGG-19, and VGG-16. The results show that the customized model achieves an accuracy of 99.5% and performs optimally according to various performance metrics. The model is implemented on an edge AI device, such as Raspberry Pi, to bring it to practical use. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2025.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Detecting data manipulation errors in android applications using scene-guided exploration

引用

Empirical Software Engineering 2025年第4期30卷 1-38页

作者： Liu, Shuqi Zhou, Yu Yang, Wenhua Chen, Taolue Gall, Harald College of Computer Science and Technology Nanjing University of Aeronautics and Astronautics Nanjing China School of Computing and Mathematical Sciences Birkbeck University of London London United Kingdom Department of Informatics University of Zurich Zurich Switzerland

Data manipulation functionalities (DMFs) refer to operations such as create, read, update and delete which are crucial for maintaining the integrity of Android application (app) data. Existing testing techniques often fail to explicitly verify DMFs in combination, having limited capability to uncover non-crashing data manipulation errors (DMEs). In this paper, we propose a novel approach SceneData, which utilizes a scene-guided exploration strategy to effectively detect DMEs in Android apps. Particularly, we model the UI pages as scenes and present a scene transition graph (SceneTG) to capture the relation between scenes. By utilizing SceneTG, we explore app states to thoroughly validate DMFs across different scenes. We evaluate SceneData on 17 popular real-world apps, and SceneData discovers 25 previously unknown bugs in their latest releases, 21 of which are non-crashing DMEs. The experiments also show that a significant fraction (15/21) of these bugs cannot be detected by the state-of-the-art techniques. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2025.

关键词： Android apps Model-based testing Non-crashing functional bugs Scene-based exploration

来源：评论

学校读者我要写书评

暂无评论

Optimizing VPN Security in Financial Services: Implementing 256-Bit Encryption and Tunneling Protocols for Secure Data Transmission 19

Optimizing VPN Security in Financial Services: Implementing ...

引用

19th International Conference on Ubiquitous Information Management and Communication, IMCOM 2025

作者： Ilias, Irma Syarlina Che Sameon, Sera Syarmila Zulkifli, Muhammad Zaqwan Ezzad Salleh, Mohammad Zahir Mat Malaysian Institute of Information Technology Universiti Kuala Lumpur Kuala Lumpur Malaysia College of Computing and Informatics Universiti Tenaga Nasional Selangor Malaysia CyberSecurity RawSec Kuala Lumpur Malaysia

ISBN: (纸本)9798331507817

This study examines the essential importance of security in contemporary network settings, emphasizing the increasing dependence on Virtual Private Networks (VPNs) for safe data transfer. The emphasis is on assessing the security attributes of VPNs to enhance existing network systems. This study specifically contrasts the performance and efficacy of two tunneling protocols—Layer 2 Tunneling Protocol version 3 with IPsec (L2TPv3/IPsec) and Generic Routing Encapsulation (GRE) utilizing IPsec—in a site-to-site connection. The study offers an economical solution by leveraging current infrastructure, minimizing the requirement for extra gear like antennas and rented space. Moreover, the study follows the National Institute of Standards and Technology (NIST) security protocols, specifically designed for the financial services sector, guaranteeing a secure and cost-effective approach for data transfer. Initial findings indicate that the best arrangement of VPN tunnels improves security in financial services networks, providing an effective method for protecting direct connections. © 2025 IEEE.

关键词： Cost effectiveness

来源：评论

学校读者我要写书评

暂无评论

DR3DH: A DoS resistant extended triple Diffie–Hellman for mobile edge networks

引用

Computer Networks 2025年 265卷

作者： Altaweel, Ala Bouridane, Ahmed Department of Computer Engineering College of Computing and Informatics University of Sharjah Sharjah United Arab Emirates

Mobile Edge Networks (MENs) are wireless networks that establish "close-to-end-users-clouds" to provide storage, computation, and communication services. MENs employ the Extended Triple Diffie–Hellman (X3DH) protocol to establish secret keys that provide end-to-end encryption among their nodes. X3DH relies on a cloud server that stores the public keys of the nodes to address the asynchronous scenarios when they operate offline. The cloud server in X3DH protocol represents a single point of failure (SPoF), as the exchange of secure keys is disrupted if the server fails or is compromised. Moreover, a malicious attacker, Eve, can launch various DoS attacks against the MEN nodes (e.g., packet dropping or packet modification) and wireless links (e.g., jamming) to disrupt the execution of X3DH during secure key exchange. This paper proposes the DoS-Resistant-X3DH (DR3DH) protocol to resolve the SPoF of the cloud server in X3DH and to address packet dropping, packet modification, and jamming DoS attacks. DR3DH employs Reed–Solomon Erasure Coding to encode the X3DH public keys into fragments that are distributed into a set of storage nodes. These storage nodes, which are determined by an Integer Linear Programming (ILP) solver, are optimum. That is, they have lowest probabilities to launch packet dropping/modification attacks and lowest probabilities that the wireless links toward them are jammed. The authors evaluated DR3DH through trace-driven simulations in MATLAB using two mobility traces and a proof-of-concept implementation in Java. The authors also compared DR3DH with two approaches that employ Random and Greedy policies when selecting the storage nodes for keys fragments. The results demonstrate the resistance of DR3DH to DoS attacks, achieving a high success rate that outperforms the Random and Greedy approaches by between 13.6% and 83.0%. Additionally, the results confirm the feasibility of DR3DH under different mobility models and node speeds, as demonstrated by

关键词： Packet loss

来源：评论

学校读者我要写书评

暂无评论

Leveraging Game Theory and XAI for Data Quality-Driven Sample and Client Selection in Trustworthy Split Federated Learning

引用

IEEE Transactions on Consumer Electronics 2025年

作者： Tariq, Asadullah Sallabi, Farag Serhani, Mohamed Adel Qayyum, Tariq Barka, Ezedin S. United Arab Emirates University College of IT United Arab Emirates College of Computing and Informatics UoS United Arab Emirates

In Federated Learning (FL) systems, clients share updates derived from their local data with the server while maintaining privacy. The server aggregates these updates to refine the global model. However, not all client data may be relevant to the learning objective, and incorporating updates from irrelevant data can harm the model's performance. The selection of training samples significantly impacts model performance, as datasets with errors, skewed distributions, or low diversity can lead to inaccurate and unstable models. To address these issues, in this paper, we proposed a data quality evaluation model to assess the quality of datasets in FL systems. This model aims to dynamically select high-quality data samples for FL training by utilizing intrinsic and contextual data quality dimensions. Additionally, we designed an importance-based interpretable feature selection model and a data quality-based dynamic client selection model that employs Nash equilibrium and joint differential privacy (DP). This approach encourages clients with high-quality data to participate in FL training, thereby improving the overall quality of the training process. We leverage the concept of Split FL to efficiently distribute the model training between the client and server. Our comprehensive approach ensures the selection of high-quality data and features while motivating the participation of reliable clients, ultimately leading to improved performance and stability of FL models. Extensive experimental results show that our proposed model significantly outperforms baseline and comparative schemes. © 1975-2011 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

RT-PPS: Real-time privacy-preserving scheme for cloud-hosted IoT data

引用

Journal of High Speed Networks 2025年第1期31卷 71-89页

作者： Elhoseny, Mohamed Riad, Khaled College of Computing and Informatics University of Sharjah Sharjah27272 United Arab Emirates Faculty of Computers and Information Mansoura University Mansoura35516 Egypt Computer Science Department College of Computer Sciences & Information Technology King Faisal University Al-Ahsa 31982 Saudi Arabia Mathematics Department Faculty of Science Zagazig University Zagazig44519 Egypt

The Internet of Things (IoT) is a rapidly growing network of devices that can communicate with each other and with cloud-based services. These devices generate vast amounts of data that can be used to provide valuable insights into user behavior, environmental conditions, and other important factors. However, as this data is collected and processed by cloud-hosted services, there is a growing concern about privacy and security. Without adequate protection, sensitive information could be exposed to hackers or other malicious actors, putting both individuals and organizations at risk. To address this challenge, real-time privacy-preserving techniques can be used to protect IoT data without compromising its value. This paper introduces an efficient Real-time privacy-preserving scheme (RT-PPS) for cloud-hosted IoT data. RT-PPS employs multi-authority attribute-based encryption on a hybrid cloud environment to keep data secure and private, while still allowing it to be processed and analyzed by cloud-hosted services. RT-PPS has efficient response time and resource consumption, which gives it the ability to handle a huge number of concurrent users at the same time without notable delay. The proposed RT-PPS has been validated through extensive experimental evaluation on a variety of configurations. Moreover, the proposed scheme has been computationally compared with the state-of-the-artwork. RT-PPS has shown excellent performance, effectiveness, and efficiency. The RT-PPS encryption time for a 1 GB dataset while considering 1024 slices is approximately 1000 ms. Also, the RT-PPS decryption time for a 1 GB ciphertext while considering 1024 slices are approximately 235 ms. Finally, RT-PPS is proven secure against any polynomial-time attacks and their variations that have at most a negligible advantage in the introduced security model. Moreover compared to most of the state-of-the-artwork, RT-PPS reduced the ciphertext size and lowered the computations in the encryption, key g

关键词： Ciphertext

来源：评论

学校读者我要写书评

暂无评论

Drone-Based Public Surveillance Using 3D Point Clouds and Neuro-Fuzzy Classifier

引用

Computers, Materials & Continua 2025年第3期82卷 4759-4776页

作者： Yawar Abbas Aisha Ahmed Alarfaj Ebtisam Abdullah Alabdulqader Asaad Algarni Ahmad Jalal Hui Liu Faculty of Computing and AI Air UniversityIslamabad44000Pakistan Department of Information Systems College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityRiyadh11671Saudi Arabia Department of Information Technology College of Computer and Information SciencesKing Saud UniversityRiyadh12372Saudi Arabia Department of Computer Sciences Faculty of Computing and Information TechnologyNorthern Border UniversityRafha91911Saudi Arabia Department of Computer Science and Engineering College of InformaticsKorea UniversitySeoul02841Republic of Korea Cognitive Systems Lab University of BremenBremen28359Germany

Human Activity Recognition(HAR)in drone-captured videos has become popular because of the interest in various fields such as video surveillance,sports analysis,and human-robot ***,recognizing actions from such videos poses the following challenges:variations of human motion,the complexity of backdrops,motion blurs,occlusions,and restricted camera *** research presents a human activity recognition system to address these challenges by working with drones’red-green-blue(RGB)*** first step in the proposed system involves partitioning videos into frames and then using bilateral filtering to improve the quality of object foregrounds while reducing background interference before converting from RGB to grayscale *** YOLO(You Only Look Once)algorithm detects and extracts humans from each frame,obtaining their skeletons for further *** joint angles,displacement and velocity,histogram of oriented gradients(HOG),3D points,and geodesic Distance are *** features are optimized using Quadratic Discriminant Analysis(QDA)and utilized in a Neuro-Fuzzy Classifier(NFC)for activity ***-world evaluations on the Drone-Action,Unmanned Aerial Vehicle(UAV)-Gesture,and Okutama-Action datasets substantiate the proposed system’s superiority in accuracy rates over existing *** particular,the system obtains recognition rates of 93%for drone action,97%for UAV gestures,and 81%for Okutama-action,demonstrating the system’s reliability and ability to learn human activity from drone videos.

关键词： Activity recognition geodesic distance pattern recognition neuro fuzzy classifier

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：