检索结果-内蒙古大学图书馆

Trustworthy deep learning for encrypted traffic classification

Soft computing 2025年第2期29卷 645-662页

作者： Li, Zheng Liu, Yanbei Zhang, Changqing Shan, Wanjin Zhang, Haifeng Zhu, Xiaoming School of Electronic and Information Engineering Tiangong University Tianjin300387 China School of Life Sciences Tiangong University Tianjin300387 China Tianjin Key Laboratory of Optoelectronic Detection Technology and Systems Tianjin300387 China College of Intelligence and Computing Tianjin University Tianjin300072 China Science and Technology on Communication Networks Laboratory Shijiazhuang050081 China The 54th Research Institute of China Electronics Technology Group Corporation Shijiazhuang050081 China

Network traffic classification refers to the identification of collected network traffic data of various applications, which is widely used in research fields such as network resource allocation, traffic scheduling and intrusion detection systems. With the widespread application of encryption technology in the network, encrypted traffic classification has become a hot research topic. At present, most existing methods only focus on the accuracy of network traffic classification. Yet, few work studies the reliability of the classification model, which plays an important role in network regulation and network security. In this paper, we propose a novel traffic classification method based on trustworthy deep learning model, which can effectively improve the reliability of encrypted traffic classification models by correcting the confidence of model output. Specifically, we firstly perform data preprocessing on the original network traffic, and then adopt a ConvNet for feature learning and a ClassifyNet for traffic classification in the initial stage. At the same time, we utilize a trustworthy confidence criterion to design a ConfidNet trained according to the probability of the true class. The ConfidNet can provide a reliable confidence measure for the prediction of the classification model. Finally, we demonstrate the effectiveness of our framework through comprehensive experiments on two benchmark datasets ISCX VPN-nonVPN and USTC-TFC2016, and show that our method can improve the reliability of the classification model and has a good ability to identify misclassified samples compared with state-of-the-art methods. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2025.

关键词： Network intrusion

来源：评论

学校读者我要写书评

暂无评论

An Ensemble Framework for Dropout Prediction in Online Learning

An Ensemble Framework for Dropout Prediction in Online Learn...

引用

2022 IEEE International Conference on Knowledge Engineering and Communication systems, ICKES 2022

作者： Srinivasan, Sruthi Dewan, M. Ali Akber School of Computer Science and Engineering Vellore Institute of Technology Chennai India School of Computing and Information Systems Athabasca University Faculty of Sceinece and Technology AB Canada

ISBN: (纸本)9781665456371

Online learning has gained traction over recent years, especially since online education has become more widespread. However, it comes with its own set of challenges of which high dropout is still a major one. Identifying at-risk learners at an early stage is pivotal to offering personalized attention that can potentially prevent them from dropping out from the online courses. This work proposes two methods to analyze students' progress in an online course and subsequently identify dropout prone students. The first method performs fusion of course activity features by concatenating previous weeks' features before training. The second method extracts course activity features from the start date of the courses to a current week instead of concatenating as the first method does. A set of machine learning models and an ensemble framework were trained and tested on these two types of feature sets. On evaluating the models, the benchmark dataset KDDCup15 has been used, where the first method yielded an F1-score 91% while the second method yielded a score 92%. It was observed that both feature fusion methods produce comparable results although we expected that the concatenation of the features over time would produce better results. We also found that using features over a longer duration of time can help in achieving better performance. Further, ensemble model consistently outperformed the base classifiers. © 2022 IEEE.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Automated Arabic Text Classification Using Hyperparameter Tuned Hybrid Deep Learning Model

引用

Computers, Materials & Continua 2023年第3期74卷 5447-5465页

作者： Badriyya B.Al-onazi Saud S.Alotaib Saeed Masoud Alshahrani Najm Alotaibi Mrim M.Alnfiai Ahmed S.Salama Manar Ahmed Hamza Department of Language Preparation Arabic Language Teaching InstitutePrincess Nourah bint Abdulrahman UniversityP.O.Box 84428Riyadh11671Saudi Arabia Department of Information Systems College of Computing and Information SystemUmm Al-Qura UniversitySaudi Arabia Department of Computer Science College of Computing and Information TechnologyShaqra UniversityShaqraSaudi Arabia Prince Saud AlFaisal Institute for Diplomatic Studies Saudi Arabia Department of Information Technology College of Computers and Information TechnologyTaif UniversityTaif P.O.Box 11099Taif21944Saudi Arabia Department of Electrical Engineering Faculty of Engineering&TechnologyFuture University in EgyptNew Cairo11845Egypt Department of Computer and Self Development Preparatory Year DeanshipPrince Sattam bin Abdulaziz UniversityAlKharjSaudi Arabia

The text classification process has been extensively investigated in various languages,especially *** classification models are vital in several Natural Language Processing(NLP)*** Arabic language has a lot of *** instance,it is the fourth mostly-used language on the internet and the sixth official language of ***,there are few studies on the text classification process in Arabic.A few text classification studies have been published earlier in the Arabic *** general,researchers face two challenges in the Arabic text classification process:low accuracy and high dimensionality of the *** this study,an Automated Arabic Text Classification using Hyperparameter Tuned Hybrid Deep Learning(AATC-HTHDL)model is *** major goal of the proposed AATC-HTHDL method is to identify different class labels for the Arabic *** first step in the proposed model is to pre-process the input data to transform it into a useful *** Term Frequency-Inverse Document Frequency(TF-IDF)model is applied to extract the feature ***,the Convolutional Neural Network with Recurrent Neural Network(CRNN)model is utilized to classify the Arabic *** the final stage,the Crow Search Algorithm(CSA)is applied to fine-tune the CRNN model’s hyperparameters,showing the work’s *** proposed AATCHTHDL model was experimentally validated under different parameters and the outcomes established the supremacy of the proposed AATC-HTHDL model over other approaches.

关键词： Hybrid deep learning natural language processing arabic language text classification parameter tuning

来源：评论

学校读者我要写书评

暂无评论

Novel vision transformer and data augmentation technique for efficient detection of monkeypox disease

引用

Multimedia Tools and Applications 2024年 1-19页

作者： Alarfaj, Aisha Ahmed Ahmad, Salman Hakeem, Abeer M. Alabdulqader, Ebtisam Abdullah PERO, Chiara Alsubai, Shtwai Innab, Nisreen Ashraf, Imran Department of Information Systems College of Computer and Information Sciences Princess Nourah bint Abdulrahman University P.O. Box 84428 Riyadh11671 Saudi Arabia Post Graduate Resident Department of Urology Allama Iqbal Teaching Hospital Dera Ghazi Khan Pakistan Department of Information Technology Faculty of Computing and Information Technology King Abdulaziz University Jeddah Saudi Arabia Department of Information Technology College of Computer and Information Sciences King Saud University P. O. Box 800 Riyadh11421 Saudi Arabia Department of Management & amp Innovation Systems University of Salerno Via Giovanni Paolo II 132 Fisciano Salerno Italy Department of Computer Science College of Computer Engineering and Sciences Prince Sattam bin Abdulaziz University P.O. Box 151 Al-Kharj11942 Saudi Arabia Department of Computer Science and Information Systems College of Applied Sciences AlMaarefa University Diriyah Riyadh13713 Saudi Arabia Department of Information and Communication Engineering Yeungnam University Gyeongsan Korea Republic of

Recent technological advancements have paved the way for the optimization of medical processes, particularly automated disease detection. Moreover, the adoption of machine learning (ML) has greatly helped in automating disease detection. Such approaches can detect various diseases early, enabling timely treatment to save countless lives. Early and accurate diagnosis is very important for diseases like monkeypox, to curb its spread. Monkeypox is a viral disease caused by double-stranded DNA and can be transmitted through close contact with infected humans or animals. It’s early identification and accurate lesion diagnosis are critical to contain the disease. This study proposes an automated approach to optimize the diagnosis of monkeypox disease using a novel vision transformer, which is utilized due to its effectiveness for feature extraction. The Proposed approach’s efficiency and accuracy are tested on a public benchmark dataset comprising a variety of skin lesions of different ages and genders. In addition, data augmentation involves rotation, scaling, and flipping thereby enhancing the density of the training data set for better generalization of ML models. Experiments involve binary, as well as, multi-class classification. For the binary class, the proposed model achieves an accuracy of 97.63%, outperforming traditional ML and deep learning (DL) techniques. In the case of multi-class classification with monkeypox, measles, normal, HFMD, cowpox, and chickenpox classes, the proposed model archives an accuracy of 90.61% while precision, recall, and F1 scores are 91.39%, 89.17%, and 90.28%, respectively. Furthermore, the proposed approach shows average accuracy, precision, recall, and F1 scores of 97.54%, 96.19%, 95.16%, and 95.67%, respectively for five-fold cross-validation. Experiments demonstrate that the combination of data augmentation techniques and the vision transformer model significantly optimizes diagnostic performance. In brief, advanced DL architectur

关键词： Medical image processing

来源：评论

学校读者我要写书评

暂无评论

Image Reconstruction for Lensless Imaging Using a Phase Congruency Perception Model

Image Reconstruction for Lensless Imaging Using a Phase Cong...

引用

2023 China Automation Congress, CAC 2023

作者： Tian, Yibin Lu, Dajiang Deng, Hong Zhang, Zhiyuan Zhong, Xiaopin Wu, Zongze College of Mechatronics and Control Engineering Shenzhen University Shenzhen China Litemaze Technology R&D Division Hangzhou China School of Computing and Information Systems Singapore Management University Singapore Singapore

ISBN: (纸本)9798350303759

Lensless imaging systems capture images without the need for traditional lenses. Instead, they employ flat and thin micro-optical devices, such as masks, diffusers, Fresnel zone plates, and other diffractive optical elements, to capture information-encoded images of imaging targets, and rely on advanced computational methods to reconstruct regular images from the encoded ones. Typical image reconstructions in lensless imaging utilize compressive sensing and sophisticated iterative optimization algorithms. Recently such iterative methods have been replaced by deep neural networks. However, in both approaches the cost (or loss) functions usually are defined on certain simple mathematical distance metric, without considering the inherent perceptive properties of images. In this report, a new method is proposed to construct cost functions for image reconstruction in lensless imaging. It is based on a visual perception model of image phase congruency, a measure of the consistency of the phase information across different spatial orientations and scales in images. The proposed method is demonstrated to provide improved image reconstruction quality using more than 200 encoded images from a virtual lensless imaging system with a Fresnel zone plate, as well a small number of encoded images from a physical one with an optical diffuser. © 2023 IEEE.

关键词： Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Challenges and Opportunities for Enabling the Next Generation of Cross-Domain Dataspaces 2

Challenges and Opportunities for Enabling the Next Generatio...

引用

2nd International Workshop on Semantics in Dataspaces, SDS 2024

作者： Deshmukh, Rohit A. Collarana, Diego Gelhaar, Joshua Theissen-Lipp, Johannes Lange, Christoph Arnold, Benedikt T. Curry, Edward Decker, Stefan Fraunhofer Institute for Applied Information Technology FIT Sankt Augustin Germany RWTH Aachen University Aachen Germany Fraunhofer Institute for Software and Systems Engineering ISST Dortmund Germany Insight Centre for Data Analytics University of Galway Galway Ireland

Dataspaces are regarded as a standardized solution for sharing data in a trusted way. However, providing and sharing high-quality data across dataspaces poses several scientific and technical challenges, opening new research avenues. Developing governance models and technologies for supporting cross-domain integration of data and services from the existing single-domain dataspaces represents a significant challenge. In this vision paper, we discuss the challenges for enabling next-generation dataspaces and propose an innovative approach that aims at developing a vision and identifying requirements and building blocks for next-generation dataspaces, followed by defining a roadmap and practical migration paths for the existing dataspaces towards the next-generation dataspaces. © 2024 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).

关键词： Interoperability

来源：评论

学校读者我要写书评

暂无评论

Adaptive and Accessible User Interfaces for Seniors Through Model-Driven Engineering

arXiv

引用

arXiv 2025年

作者： Wickramathilaka, Shavindra Grundy, John Madampe, Kashumi Haggag, Omar Department of Software Systems and Cybersecurity Faculty of Information Technology Monash University Wellington Road ClaytonVIC3800 Australia

The use of diverse apps among senior users is increasing. However, despite their diverse age-related accessibility needs and preferences, these users often encounter apps with significant accessibility barriers. Even in the best-case scenarios, they are provided with one-size-fits-all user interfaces that offer very limited personalisation support. To address this issue, we describe AdaptForge, a novel model-driven engineering (MDE)-based approach to support sophisticated adaptations of Flutter app user interfaces and behaviour based on the age-related accessibility needs of senior users. We explain how AdaptForge employs Domain-Specific Languages to capture seniors’ context-of-use scenarios and how this information is used via adaptation rules to perform design-time modifications to a Flutter app’s source code. Additionally, we report on evaluations conducted with real-world Flutter developers to demonstrate the promise and practical applicability of AdaptForge, as well as with senior end-users using our adapted Flutter app prototypes. © 2025, CC BY-NC-ND.

关键词： Application programs

来源：评论

学校读者我要写书评

暂无评论

Video-Based Jump Height Estimation in Athletic Performance Testing

Video-Based Jump Height Estimation in Athletic Performance T...

引用

2023 IEEE International Workshop on Sport, technology and Research, STAR 2023

作者： Lokas, Ivan Vasilj, Matea Skender, Sven Vucetic, Vlatko Mihaldinec, Hrvoje Dzapo, Hrvoje Ultrax Technologies Software Development Zagreb Croatia Human Performance Laboratory Faculty of Kinesiology Zagreb Croatia Hardware Development Ultrax Technologies Zagreb Croatia Faculty of Electrical Engineering and Computing Department of Electronic Systems and Information Processing Zagreb Croatia

ISBN: (纸本)9798350316056

Jump height estimation is well-studied problem in sports science with a variety of methods proposed over the years. This paper introduces a portable, simple-to-setup and a cost-effective video-based jump height estimation method, and compares it with existing traditional methods that use expensive and specialized equipment. To enable more accurate comparison between different systems, a custom-designed Inertial Measurement Unit (IMU) was used as an additional sensor input. The analysis was performed on eleven female physically active handball players. The dataset consists of three most common vertical jump variations: squat jump, counter-movement jump and counter-movement jump with free arms. The analysis show that our video-based approach achieves comparable results to the traditional methods. © 2023 IEEE.

关键词： Cost effectiveness

来源：评论

学校读者我要写书评

暂无评论

GemmaWithLoRA: A New Approach to Click Fraud Detection

GemmaWithLoRA: A New Approach to Click Fraud Detection

引用

2024 IEEE International Conference on Communications, computing, Cybersecurity and Informatics, CCCI 2024

作者： Dai, Bingzhou Wang, Lei Zhao, Xiaoyong Guo, Yu Obaidat, Mohammad S. School of Information Management Beijing Information Science & Technology University Beijing China Information Systems Institute Beijing Advanced Innovation Center for Materials Genome Engineering Beijing Information Science & Technology University Beijing China Beijing Advanced Innovation Center for Materials Genome Engineering and Shunde Innovation School University of Science and Technology Beijing China King Abdullah II School of Information Technology University of Jordan School of Engineering Amity University School of Computing SRM University Amman Jordan

ISBN: (纸本)9798350349832

With the rapid development of artificial intelligence technology, pre-Trained large language models (LLMs) have demonstrated their powerful application potential in various fields. Click fraud is a serious issue in the digital advertising ecosystem, especially in the pay-per-click (PPC) model, where fraudulent activities significantly undermine the effectiveness of ads and pose security risks. While many traditional fraud detection methods have achieved relatively good accuracy, they still face poor generalization and extensive feature engineering issues. This paper explores using LLMs for click fraud detection by leveraging the powerful feature extraction capabilities and superior generalization performance of the Gemma-2b model, combined with LoRA fine-Tuning technology. Extensive experiments on the TalkingData2017 dataset showed that accuracy was achieved at 83°/0, demonstrating the effectiveness of LLMs in click fraud detection tasks. © 2024 IEEE.

关键词： Crime

来源：评论

学校读者我要写书评

暂无评论

Enhancing Secure Development in Globally Distributed software Product Lines: A Machine Learning-Powered Framework for Cyber-Resilient Ecosystems

引用

Computers, Materials & Continua 2024年第6期79卷 5031-5049页

作者： Marya Iqbal Yaser Hafeez Nabil Almashfi Amjad Alsirhani Faeiz Alserhani Sadia Ali Mamoona Humayun Muhammad Jamal University Institute of Information Technology PMAS-Arid Agriculture UniversityRawalpindiPakistan Department of Software Engineering College of Computer and Information SciencesJouf UniversityAl Jouf72388Saudi Arabia Department of Computer Science College of Computer and Information SciencesJouf UniversityAl Jouf72388Saudi Arabia Department of Computer Engineering&Networks College of Computer and Information SciencesJouf UniversityAl Jouf72388Saudi Arabia Department of Information Systems College of Computer and Information SciencesJouf UniversitySakakaAl Jouf72388Saudi Arabia Department of Mathematics Pir Mehr Ali Shah Arid Agriculture UniversityRawalpindiPakistan

Embracing software product lines(SPLs)is pivotal in the dynamic landscape of contemporary software ***,the flexibility and global distribution inherent in modern systems pose significant challenges to managing SPL variability,underscoring the critical importance of robust cybersecurity *** paper advocates for leveraging machine learning(ML)to address variability management issues and fortify the security of *** the context of the broader special issue theme on innovative cybersecurity approaches,our proposed ML-based framework offers an interdisciplinary perspective,blending insights from computing,social sciences,and ***,it employs ML for demand analysis,dynamic feature extraction,and enhanced feature selection in distributed settings,contributing to cyber-resilient *** experiments demonstrate the framework’s superiority,emphasizing its potential to boost productivity and security in *** digital threats evolve,this research catalyzes interdisciplinary collaborations,aligning with the special issue’s goal of breaking down academic barriers to strengthen digital ecosystems against sophisticated attacks while upholding ethics,privacy,and human values.

关键词： Machine Learning variability management cybersecurity digital ecosystems cyber-resilience

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：