检索结果-内蒙古大学图书馆

Analyzing darknet traffic through machine learning and neucube spiking neural networks

Intelligent and Converged Networks 2024年第4期5卷 265-283页

作者： Iman Akour Mohammad Alauthman Khalid M.O.Nahar Ammar Almomani Brij B.Gupta Information Systems Department College of Computing&InformaticsUniversity of SharjahSharjahUnited Arab Emirates Department of Information Security Faculty of Information TechnologyUniversity of PetraAmman 11196Jordan the Faculty of Computer Studies Arab Open UniversityRiyadh 11681Saudi Arabia Department of Information Technology Al-Huson University CollegeAl-Balqa Applied UniversityIrbid 19117Jordan the Department of Computer Information Science Higher Colleges of TechnologySharjahUnited Arab Emirates the Department of Computer Science and Information Engineering Asia UniversityTaichung 413ChinaSymbiosis Centre for Information TechnologySymbiosis International(Deemed University)Pune 412115India Centre of Inter-disciplinary Research&Innovation University of Petroleum and Energy StudiesDehradun 248007India

The rapidly evolving darknet enables a wide range of cybercrimes through anonymous and untraceable communication *** detection of clandestine darknet traffic is therefore critical yet immensely *** research demonstrates how advanced machine learning and specialized deep learning techniques can significantly enhance darknet traffic analysis to strengthen *** diverse classifiers such as random forest and naïve Bayes with a novel spiking neural network architecture provides a robust foundation for identifying concealed *** on the CIC-Darknet2020 dataset establishes state-of-the-art results with 98%accuracy from the random forest model and 84.31%accuracy from the spiking neural *** pioneering application of artificial intelligence advances the frontiers in analyzing the complex characteristics and behaviours of darknet *** proposed techniques lay the groundwork for improved threat intelligence,real-time monitoring,and resilient cyber defense systems against the evolving landscape of cyber threats.

关键词： darknet traffic machine learning deep learning spiking neural network PCA

来源：评论

学校读者我要写书评

暂无评论

Jupiter: a modern federated learning platform for regional medical care

引用

Science China(Information Sciences) 2021年第10期64卷 152-165页

作者： Ju XING Jiadong TIAN Zexun JIANG Jiali CHENG Hao YIN Department of Automation Tsinghua University School of Cyberspace Security Beijing University of Posts and Telecommunications Department of Computer Science and Technology Tsinghua University College of Engineering Northeastern University

With the emergence of AI technologies, intrinsic value of data is released and takes tremendous effects on numerous industries. In the context of regional medical care, data sharing and cooperating is in high demand, which can bring both financial and societal benefits. At present, however, medical data are locked inside medical facilities owing to legal risks and economic considerations. How to bring AI technologies into full play under this circumstance is a big challenge. In this paper, we propose Jupiter, an easy-to-use,secure, and high-performance platform for federated machine learning. Jupiter constructs a secure and highperformance aggregator cluster with SGX to efficiently aggregate the encrypted model parameters. Jupiter employs a stateful design to cooperate with medical facilities in regional medical systems with a fixed network connection. By providing an innovative programming abstraction, Jupiter makes model development more friendly to developers. The experiments show that with a low memory footprint, the throughput of a single node on an ordinary PC can reach 300 MB/s(with slice size fixed to 64 KB), and the aggregation primitive we built can process 11 k aggregations per second.

关键词： federated learning programming abstraction SGX high-performance

来源：评论

学校读者我要写书评

暂无评论

Outlier-aware Cross-Market Product Recommendation

Outlier-aware Cross-Market Product Recommendation

引用

International Conference on Big Data and Smart Computing (BIGCOMP)

作者： HyeoungGuk Kang Donghoon Lee Hyunsouk Cho Department of Cyber Security Ajou University Department of Software and Computer Engineering Ajou University Department of Artificial Intelligence Ajou University

Cross Market Recommendation (CMR) is a method of recommending in a resource-scarce market by using modelagnostic meta-learning. Generally, more interactions give more clues to identify the user preferences, so CF performs better with outlier users (who have more item interactions) than normal users. However, constructing each adapt batch set (support set) and evaluation batch set (query set) for meta-learning in CMR causes the model to underfit in outlier users. We aim at this phenomenon and propose a new hybrid strategy to solve this problem. By simply combining MAML and CF to target general users and outliers, respectively. We also validate our method with the benchmark dataset and the proposed model shows better performance compared to the original model.

关键词： Adaptation models Computational modeling Neural networks Big Data Benchmark testing

来源：评论

学校读者我要写书评

暂无评论

Contrastive Learning-Based Generic Audio-to-Lyrics Alignment 21

Contrastive Learning-Based Generic Audio-to-Lyrics Alignment

引用

21st IEEE International Symposium on Parallel and Distributed Processing with Applications, 13th IEEE International Conference on Big Data and Cloud Computing, 16th IEEE International Conference on Social Computing and Networking and 13th International Conference on Sustainable Computing and Communications, ISPA/BDCloud/SocialCom/SustainCom 2023

作者： Jin, Zhenkun Huang, Shuai Nie, Xin Zhou, Xinlei Yi, Yuanyuan Zhou, Gefei Wuhan Business UIniversity Wuhan China Wuhan Research Institute of Posts and Telecommunications Wuhan China Hubei Key Laboratory of Distributed System Security Wuhan China Hubei Engineering Research Center on BigData Security Wuhan China Huazhong University of Science and Technology School of Cyber Science and Engineering Wuhan China Huazhong University of Science and Technology School of Journalism and Information Communication Wuhan China Huazhong University of Science and Technology School of Computer Science and Technology Wuhan China

ISBN: (纸本)9798350329223

In this research, we tackled the complex task of aligning audio and lyrics content automatically. This task entails the precise matching of lyrics with the corresponding audio segments in songs, necessitating the coordination of two distinct modalities: text and audio. To surmount this challenge, we employed a contrastive learning approach that generates cross-modal embeddings capable of establishing connections between the audio and text domains. Furthermore, we adopted a hierarchical and cascading design. Initially, the system predicts synchronization at the sentence level and subsequently refines the alignment at the word level. This design equips the system to handle extended sequences effectively, as cross-correlations leverage secondary memory that corresponds to sequence length. Through this innovative approach, we developed a trainable end-to-end system that can harness weakly annotated training data, concurrently learning a robust text model optimized for alignment tasks. Our experiments demonstrated substantial enhancements in average alignment accuracy when utilizing our proposed system, underscoring its robustness compared to previous state-of-the-art models. © 2023 IEEE.

关键词： Alignment

来源：评论

学校读者我要写书评

暂无评论

Efficient Neuroimaging Data security and Encryption Using Pixel-Based Homomorphic Residue Number System

引用

SN computer Science 2023年第6期4卷 1-10页

作者： Usman, Lateef Opeyemi Muniyandi, Ravie Chandren Usman, Morufat Adebola Research Centre for Cyber Security Faculty of Information Science and Technology Universiti Kebangsaan Malaysia Selangor UKM Bangi 43600 Malaysia Department of Computer Science Tai Solarin University of Education P.M.B. 2118 Ogun State Ijagun Ijebu-Ode Nigeria

In recent times, there has been an increasing attention in designing a homomorphic privacy-preserving classification method for neuro-images based on the residue number system (RNS) and deep CNN models. This article presents the RNS homomorphic encryption system for neuro-images and evaluates its security efficiency with respect to moduli set { 2 ⁿ- 1 , 2 ⁿ, 2 ⁿ⁺¹- 1 } . The efficiency of the proposed system is evaluated through the application of three metrics, namely visual inspection, encoding analysis, and security analysis. The analysis demonstrates that the proposed RNS scheme is a fully homomorphic encryption (FHE) scheme that can encrypt and decrypt neuroimages without compromising any essential neural biomarker features. Additionally, the scheme is robust against statistical attacks like histogram, brute force, correlation coefficient, and key sensitivity. Thus, the proposed RNS-FHE scheme can be utilized for any neuroimaging dataset and is appropriate for the design of homomorphic privacy-preserving methods when compared to the current state-of-the-art. In summary, the contribution and novelty of this study is the development of a systematic RNS-FHE privacy-preserving approach for efficient neuroimaging dataset homomorphic encryption. © 2023, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd.

关键词： Cipher images Homomorphic encryption Neural biomarkers Neuroimages Residue number system

来源：评论

学校读者我要写书评

暂无评论

Utilizing Machine Learning Techniques for Enhanced Predictive Maintenance in the Manufacturing Sector

Utilizing Machine Learning Techniques for Enhanced Predictiv...

引用

International Conference on Networking, Embedded and Wireless Systems (ICNEWS)

作者： Rashmi N. Wadibhasme Mallireddy Naresh G.Vikram Arun Sankar V V Suresh P F.Jermina Department of Information Technology YCCE Nagpur Maharashtra India Department of Computer Science Engineering Aditya College of Engineering and Technology Karunya School of Management Karunya Institute of Technology and Sciences Coimbatore Department of Mechanical Engineering Karpagam College of Engineering Coimbatore Mechanical Engineering Muthayammal Engineering College Department of Cyber Security Karpagam College of Engineering Coimbatore

ISBN: (数字)9798350367461

ISBN: (纸本)9798350367478

The contributions of machine learning algorithms to predictive maintenance systems, such as classification, regression, and anomaly detection, are investigated to clarify their functions in forecasting equipment malfunctions and optimizing maintenance schedules. Notwithstanding the potential for profound change, specific obstacles including the quality of data, the ease of interpretation, and the financial implications of implementation require careful deliberation. This paper highlights the concrete advantages gained from the implementation of predictive maintenance by employing case studies and numerical analysis. These benefits include decreased equipment outage, improved equipment reliability, and lower maintenance expenses. The results emphasize the critical nature of predictive maintenance in the manufacturing sector, shedding light on its potential to transform processes and guarantee ongoing efficiency. With the increasing adoption of digitalization and analytics across industries, predictive maintenance emerges as a critical element in enhancing manufacturing sector efficiency and competitiveness. This highlights the ongoing imperative for innovative and adaptable practices that strive for operational excellence. In performance comparisons for predictive maintenance, Neural Networks achieved the highest metrics with 96.3% accuracy, 94.8% precision, 97.2% recall, and a 95.9% F1 score. Random Forests followed with 94.5% accuracy, 92.7% precision, 95.8% recall, and a 94.2% F1 score. Decision Trees showed 92.1% accuracy, 89.3% precision, 94.2% recall, and 91.7% F1 score, while Logistic Regression had the lowest performance with 85.2% accuracy, 82.6% precision, 87.5% recall, and an 84.8% F1 score.

关键词： Schedules Accuracy Machine learning algorithms Organizations Data models Manufacturing Maintenance Asset management Anomaly detection Predictive maintenance

来源：评论

学校读者我要写书评

暂无评论

AI-Driven Solutions for Cloud security Implementing Intelligent Threat Detection and Mitigation Strategies

AI-Driven Solutions for Cloud Security Implementing Intellig...

引用

Intelligent Computing and Emerging Communication Technologies (ICEC), International Conference on

作者： Rajashekar Reddy Yasani Putalpattu Muni Prasad Pattlola Srinivas N V Raja Sekhar Reddy Parag Jawarkar Vedaprada Raghunath Cloud Security Cloud Computing Cyber Security Department of Industrial Engineering University of Rome Tor Vergata Department of Computer Science and Engineering Malla Reddy Engineering College(A) Telangana India Department of IT MLR Institute of Technology India Electronics Engineering Shri Ramdeobaba College of Engineering and Management Nagpur Visveswaraya Technological University India

ISBN: (数字)9798331508432

ISBN: (纸本)9798331508449

Adapting threat detection and response procedures is essential for businesses to stay ahead of the constantly developing cyberattack landscape. Researching cutting-edge tools like XDR, SIEM, SOAR, and NDR—which shed light on the dynamic realm of detection and response systems—is crucial. Other areas of investigation include real-time monitoring and network forensics. security and threat detection have both been greatly enhanced since businesses began storing their apps and data in the cloud. In order to safeguard vulnerable network infrastructures in cloud settings against advanced attacks, traditional security procedures must be revised. Artificial intelligence (AI) is used to learn more about this problem so that threat reaction and identification can be done more quickly and accurately. Artificial intelligence has changed cloud security and threat identification, as this study shows. Because cyberattacks on cloud infrastructures and service providers are becoming more common, strong security steps that are also easy to use are still needed. This piece will talk about how cloud security and AI operations work together, with a focus on how this collaboration speeds up the time it takes to respond to incidents. The paper will also show how this relationship helps a company improve its defences and lessen the impact of security events. Organisations that want to keep up with the constantly changing danger landscape need to use cloud security and AI and understand how the two work together. This is because it is necessary to keep a security stance that is both flexible and strong.

关键词： Training Cloud computing security Companies Threat assessment Real-time systems Safety security Artificial intelligence cyberattack Business

来源：评论

学校读者我要写书评

暂无评论

Tomato Leaf Disease Detection through Machine Learning based Parallel Convolutional Neural Networks 7

Tomato Leaf Disease Detection through Machine Learning based...

引用

7th International Conference on Intelligent Computing and Control Systems, ICICCS 2023

作者： Shetty, Sahana Selvanayaki, S. Sumathi, T. Nageswari, C.Shobana Nisha Devi, K. Suganyadevi, S. FET-Jain Department of CSE-AIML Cyber Security School of CSE Jayanagar Bangalore India Saveetha Engineering College Department of Artificial Intelligence and Data Science Chennai India Dr.Mahalingam College of Engineering and Technology Department of Information Technology Pollachi India R.M.D. Engineering College Department of ECE Kavaraipettai601206 India Bannari Amman Institute of Technology Department: Artificial Intelligence and Data Science Sathyamangalam India KPR Institute of Engineering and Technology Department of ECE Coimbatore India

ISBN: (纸本)9798350397253

Agriculture is the primary economic activity in a significant number of nations. Yet, plant-related illnesses are the most difficult obstacle for farmers to overcome. The leaf is the portion of the plant that is noticed the most, but the process of segmenting an image of a sick lesion on a leaf suffers from challenges related to uneven lighting and a crowded, complicated natural environment. The combination of colour balancing and superpixel is the foundation of the approach that has been suggested as a solution to these issues. To begin, the picture that is being read in will be converted into a color-balanced image so that the effects of uneven lighting may be removed. Second, using the superpixel operation, compact areas are produced from the modified picture. An experimentally calculated threshold is then applied to the Histogram of Gradients (HOG) and colour channels of the superpixel to separate the undesirable backdrop from the image of the leaf. In order to identify a sick or contaminated picture, K-means clustering is employed. Pyramid of HOG (PHOG), an expanded version of HOG, together with Grey Level Co-occurrence Matrix (GLCM) characteristics are used to depict a sick infected portion of the body. In the end, many disease classifiers are evaluated against one another, and Random Forest (RF) is chosen as the best option. An experiment is carried out to verify the effectiveness of the proposed approach's accuracy using natural photographs obtained from a variety of farms, images downloaded from the internet, and the Plant-Village colour dataset. This is done so that the resilience of the method can be evaluated. © 2023 IEEE.

关键词： Color

来源：评论

学校读者我要写书评

暂无评论

Personalized Shopping Experiences: Leveraging Big Data Analytics in E-commerce for Enhanced Customer Engagement

Personalized Shopping Experiences: Leveraging Big Data Analy...

引用

2024 International Conference on Decision Aid Sciences and Applications, DASA 2024

作者： Raja, Abdullah Siddique, Fatima Hasan, Muhammad Zulkifl Hussain, Muhammad Zunnurain Altaf, Junaid Kousar, Shagufta University of Central Punjab Faculty of Information Technology and Computer Science Lahore Pakistan University Putra Malaysia Researcher Cyber Security Center Malaysia Bahria University Lahore Campus Department of Computer Science Lahore Pakistan Gulab Devi Educational Complex Lahore Pakistan Applied Science Research Center Applied Science Private University Amman Jordan

ISBN: (纸本)9798350369106

E-commerce systems have integrated big data analytics (BDA) as a core element, enabling personalized shopping experiences, crucial for customer loyalty and satisfaction. This research explores how intentional use of data analytics impacts customers and signals the combined effect of technology and market data in tailoring customer needs through innovation. We developed unique data mining processes from e-commerce archives to uncover patterns in customer behavior, aiming to simplify consumer analysis and identify factors influencing purchasing and loyalty barriers. Our research also tests personalization strategies on sample data to improve key performance indicators like conversion rates, average order value, and customer retention. Ethical data use and privacy are prioritized, ensuring trust and transparency between e-commerce platforms and consumers. This study is innovative, demonstrating the power of analytics to revolutionize e-commerce by prioritizing customer needs and sustainable growth. © 2024 IEEE.

关键词： Marketplaces

来源：评论

学校读者我要写书评

暂无评论

Learners’ Privacy-Preserving Scheme for Ranking Data Sensitivity in Mobile Learning System

Learners’ Privacy-Preserving Scheme for Ranking Data Sensit...

引用

Electrical and computer engineering Researches (ICECER), International Conference on

作者： Muhammad Kudu Muhammad Olayemi Mikail Olaniyi Francis Osang Ishaq Oyebisi Oyefolahan Ojeniyi Joseph Adebayo Idris Mohammed Kolo Department of Computer Science Federal University Technology Minna Nigeria Department of Cyber Security National Open University of Nigeria Abuja Nigeria Department of Computer Science National Open University of Nigeria Abuja Nigeria Africa Centre of Excellence on Technology Enhanced Learning National Open University of Nigeria Abuja Nigeria Department of Cyber Security Federal University Technology Minna Nigeria

ISBN: (数字)9798331539733

ISBN: (纸本)9798331539740

Mobile Learning System (MLS) is facing new challenges in terms of privacy, such as data collection, storage, and sharing because of the core infrastructure and network that enables cloud computing services. Most of the data exchange in Mobile Learning System (MLS) require mandatory authorisation to allow access to the learners’ information in the MLS. Therefore, this article attempts to rank learners' sensitive attributes stored in MLS. Thus, concerns about privacy breaches motivated this paper to adopt an attributes partitioning strategy into the sensitive attributes to enforce privacy during learners’ profile information access. The article adopted the informed consent phenomenon to determine and formulate learners’ data privacy attributes sensitivity using the Fuzzy Analytic Hierarchy Process (FAHP) Algorithm. Results from the implemented Learners’ Privacy Preserving (LPP) Algorithm determined normalized weights of top-five rank-selected learners’ sensitive data to include: Browsing History (1 ST , Ranked), Geolocation Data (2 ND , Ranked), IP Address (3 RD , Ranked), web Browser (4 TH , Ranked), Medical Records (5 TH , Ranked) and CGPA (10 TH , Ranked) respectively. This indicates that these five most sensitive features are at risk and require protection to prevent privacy breaches, thus ensuring privacy preservation that prevents unauthorised access to learners’ sensitive data in the mobile learning system environment. The ranking of sensitive data in this paper could serve as inspiration for future research work on mobile learning security to improve the privacy of sensitive attributes in MLS environment

关键词： Authorization Data privacy Sensitivity Geology Privacy breach Mobile learning Partitioning algorithms Browsers IP networks History

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：