检索结果-内蒙古大学图书馆

International Journal of Intelligent Networks 2025年 6卷 47-56页

作者： Chen, Han Teng, Xuyang Su, Jiajie Li, Chunhao Hu, Chang Han, Meng School of Communication Engineering Hangzhou Dianzi University Hangzhou China Department of Computer Science National University of Singapore Singapore College of Computer Science and Technology Zhejiang University Hangzhou China

In the optimization of intelligent network architecture, limited resources at each node, including edge computing devices, have posed challenges for deploying large models in performance-demanding scenarios. Knowledge distillation serves as a model compression method that extracts knowledge from a large-scale teacher model and transfers it to a more lightweight student model. Previous knowledge distillation methods mainly focus on the intermediate layers of the network. However, due to privacy protection regulations that limit data sharing and access as well as computational efficiency requirements in practical scenarios, feature-based distillation encounters challenges in practical applications. We start with logit-based distillation to address these issues, enabling students to learn more representative knowledge from the teacher's output probability distribution. Due to the structural limitations of the teacher network such as insufficient depth or width, and potential issues in the training data like noise and imbalance, the output probability distribution contains many errors. Therefore, we propose a knowledge distillation method that improves the student by correcting errors in the teacher model. Nevertheless, teacher's errors not only bring mistakes to students but also give students greater subjectivity, enabling them to break free from the limitations of the teacher. We also retain the teacher's thinking to prevent students from becoming biased on the remaining (non-target) categories while correcting teacher errors for students. Extensive experiments demonstrate that our method achieves competitive performance on multiple benchmarks without extra parameters. © 2025 The Authors

关键词： Normal distribution

来源：评论

学校读者我要写书评

暂无评论

Advanced and Secure Data Sharing Scheme with Blockchain and IPFS: A Brief Review 15

Advanced and Secure Data Sharing Scheme with Blockchain and ...

引用

15th International Conference on Computing Communication and Networking Technologies, ICCCNT 2024

作者： Mukhedkar, Moresh Kote, Prajwal Zonde, Mounesha Jadhav, Om Bhasme, Vaibhav Dawande, Nitin A. D Y Patil University School of Engineering and Technology E&tc Engineering Department Pune India D Y Patil University School of Engineering and Technology Computer Engineering Department Pune India

ISBN: (纸本)9798350370249

In the ever-evolving information age, the challenges posed by the exponential growth of data have led to innovative solutions that leverage the power of technology. This paper introduces a comprehensive data sharing platform that seamlessly integrates blockchain and InterPlanetary File System (IPFS) technologies to address the complex issues of data sharing and storage. The approach begins by establishing an alliance blockchain, employing a consensus mechanism rooted in computing power competition to maintain data integrity within the blockchain. Simultaneously, the IPFS system is harnessed for data storage, utilizing distributed storage, file splitting, and splicing techniques. The resultant data sharing platform consists of distinct modules, including blockchain, IPFS, encryption and decryption, and fast retrieval, providing end-to-end data security and accessibility. Furthermore, the paper delves into the practical implementation of this technology by examining a course sharing project that leverages IPFS and Ethereum blockchain. By utilizing IPFS as an access medium, the paper efficiently reduces storage costs and enhances storage efficiency, effectively addressing the challenge of managing extensive course data. The incorporation of Ethereum blockchain technology ensures the elimination of single points of failure and establishes a robust transaction mechanism that incentivizes course users to engage in sharing and enhances course liquidity. This innovative amalgamation of blockchain and IPFS not only resolves the data sharing conundrum but also guarantees secure, decentralized, and efficient data storage and exchange in an era defined by the relentless growth of digital information. © 2024 IEEE.

关键词： blockchain encryption ethereum IPFS security

来源：评论

学校读者我要写书评

暂无评论

Dielectric Response Analysis Characteristic of Service-Aged XLPE Cables 14

Dielectric Response Analysis Characteristic of Service-Aged ...

引用

14th International Conference on the Properties and Applications of Dielectric Materials, ICPADM 2024

作者： Thongkong, Patchara Promphanich, Wiboon Udomluksananon, Patt Jeenmuang, Siwakorn Wiangtong, Theerayod Pattanadech, Norasage School of Engineering King Mongkut's Institute of Technology Ladkrabang Department of Electrical Engineering Bangkok Thailand School of Engineering King Mongkut's Institute of Technology Ladkrabang Department of Computer Engineering Bangkok Thailand

ISBN: (纸本)9798350374605

The main insulation used in plastic underground cables is cross-linked polyethylene (XLPE). The XLPE underground cable which has been utilized in service for a long time, has experienced various stresses, or so-called TEAM stresses. Consequently, the XLPE insulation will be degraded over time and lead to insulation properties degradation. This paper represents the dielectric response analysis technique for characterizing the insulation condition of the service-aged XLPE medium voltage underground cable. The dielectric response analysis was performed on the 22 kV XLPE underground cables that have been in service for more than 10 years and were directly buried. The dielectric response analysis measurement consists of a combination of polarization and depolarization current (PDC) measurement in the time-domain and frequency-domain spectroscopy (FDS) measurement. After that, test results were analyzed to evaluate the insulation condition and compare it to the new 22 kV underground cable. The analyzed dielectric response analysis test results will be reported in this paper. © 2024 IEEE.

关键词： Cable sheathing

来源：评论

学校读者我要写书评

暂无评论

A Vulnerability Detection Method for Smart Contracts Based on Dynamic Meta Optimizer

引用

IEEE Internet of Things Journal 2025年第13期12卷 24904-24915页

作者： He, Daojing Gong, Wei Chan, Sammy Harbin Institute of Technology School of Computer Science and Technology Shenzhen China Jiangxi University of Science and Technology School of Information Engineering Jiangxi Ganzhou China City University of Hong Kong Department of Electrical Engineering Hong Kong Hong Kong

With the increasingly complex blockchain technology environment and emerging security threats, the detection and prevention of vulnerabilities in blockchain smart contracts have become crucial for ensuring the healthy development of blockchain technology and avoiding substantial economic losses. Recently developed vulnerability detection methods for smart contracts suffer from the drawbacks of insufficient feature extraction and inadequate multitask detection. This paper proposes a multifaceted learning model, DLR, based on a dual-loop architecture of meta learning to address these issues by adopting model-agnostic meta-learning techniques. This model employs a syntax analyzer for targeted feature extraction, with graph information used as a supporting tool during the initial stage of feature engineering. Combined with an improved optimizer algorithm in the inner loop, the model can effectively learn and adjust based on the specific requirements of each task. In the outer loop, the model achieves efficient learning rate adjustment for multi-task learning by integrating the adaptive learning rate of the Adam optimizer with a global adjustment using simulated annealing strategy, thereby enhancing performance across diverse tasks. Experimental results demonstrate significant improvements in detection accuracy over state-of-the-art methods for three types of vulnerabilities, with our method achieving detection accuracies of 94.40%, 93.36%, and 94.33% for reentrancy, timestamp dependence, and integer overflow vulnerabilities, respectively. © 2014 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Forensic Eye: Convolutional Neural Network-based Crime Scene Object Detection System 2

Forensic Eye: Convolutional Neural Network-based Crime Scene...

引用

2nd International Conference Computational and Characterization Techniques in engineering and Sciences, IC3TES 2024

作者： Sudha, I. Vadivel, M. Lakshmiramana, P. Kalpana, N. Kumar, Nellore Manoj Vinoth, R. Saveethe University Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Department of Computer Science Tamilnadu Chennai India Excel Engineering College Department of Information Technology Komarapalayam Namakkal India Madanapalle Institute of Technology & Science Department of Computer Science & Technology Andhra Pradesh India Psna College of Engineering and Technology Department of Computer Science and Engineering India Saveetha School of Engineering Department of Mathematics Tamilnadu Chennai India Rmk College of Engineering and Technology Department of Computer Science and Engineering Puduvoyal India

ISBN: (纸本)9798350364699

Forensic investigations rely heavily on the accurate detection and analysis of crime related objects present at crime scenes. Traditional methods of object detection often involve manual labor and are error prone process to human error. In recent years, Convolutional Neural Networks (CNNs) have shown remarkable success in various computer vision tasks, including in object detection. In this article, we introduce ForensicEye, a novel CNN-based crime scene object detection system designed to assist forensic investigators in identifying the objects related to the crime. It utilizes a state-of-the-art CNN architecture trained on a large dataset of crime scene images, enabling robust detection even in challenging lighting and environmental conditions. We evaluated the performance of the ForensicEye on benchmark datasets with various preprocessing techniques, demonstrating its superior accuracy and efficiency compared to traditional methods. Our proposed model achieved the accuracy of 96.2%, precision of 97.9%, Recall of 96.9%, F1-score of 97.4%. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

A Hybrid Approach to Dynamic Thread Allocation Driven by Workload and Task Prioritization

引用

SN computer Science 2025年第3期6卷 1-19页

作者： Abirami, K. Harini, N. Department of Computer Science and Engineering Amrita School of Computing Coimbatore Amrita Vishwa Vidyapeetham Coimbatore India

In today’s rapidly changing world, cloud service providers face numerous challenges in managing resources and meeting customer demands. To address these challenges, cloud service providers should prioritize the tasks and allocate threads efficiently. By allocating threads efficiently, service providers can optimize resource utilization and reduce response times and job completion times which in turn improves customer satisfaction. We propose a hybrid model that integrates the threshold-driven thread Balancer (TDTB) for dynamic thread allocation with a Customer Retention Model (CRM) to prioritize tasks based on customer behaviour. Our model initiates the establishment of a thread pool configured with a dynamic threshold based on workload (on-demand requests). Concurrently, a queue to handle incoming customer requests is set up alongside a CRM priority queue designed for task prioritization. Our model calculates Job duration and completion times for n priority classes. Crown consumers experience shorter durations and completion times compared to other priority classes. Tasks are completed in descending order of priority. The results of thorough experimentation clearly show that crown consumer tasks are highly satisfied. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2025.

关键词： Customer segmentation Response time Service differentiation Thread allocation Throughput Workload

来源：评论

学校读者我要写书评

暂无评论

A Yet Powerful Routing Protocol for Vehicular Ad-hoc Networks (VANETs) Utilizing Software Defined Network (SDN) and Fog Computing 8

A Yet Powerful Routing Protocol for Vehicular Ad-hoc Network...

引用

8th International Conference on Information Technology, Information Systems and Electrical engineering, ICITISEE 2024

作者： Darabkh, Khalid A. Al-Mistarihi, Mamoun F. Al-Maaitah, Mera Ismail School of Engineering The University of Jordan Computer Engineering Department Amman11942 Jordan Jordan University of Science and Technology Faculty of Engineering Electrical Engineering Department Irbid22110 Jordan

ISBN: (纸本)9798350368970

In this article, we propose a novel protocol that employs different technologies together, such as Software Defined Network (SDN) for the purpose of offering scalability, programmability, and global network information, and Fog computing for the reason of providing location services that meet V ANET standards. Additionally, we propose a novel cluster head selection algorithm for the purpose of enhancing the efficiency of the protocol by considering three different dimensions, namely, lifetime, average distance, and signal-to-interference noise ratio. The novelty of this proposed algorithm comes from reducing the number of control overheads, reducing long-distance communications, and reducing packet loss due to collisions and contentions inside each cluster, respectively. The proposed protocol has been assessed using N s3 simulator whereas its results are so promising and worth publishing. © 2024 IEEE.

关键词： Vehicular ad hoc networks

来源：评论

学校读者我要写书评

暂无评论

N-Gram Based Sarcasm Detection for News and Social Media Text Using Hybrid Deep Learning Models

引用

SN computer Science 2024年第1期5卷 163页

作者： Thaokar, Chetana Rout, Jitendra Kumar Rout, Minakhi Ray, Niranjan Kumar School of Computer Engineering KIIT Deemed to be University Odisha Bhubaneshwar India Ramdeobaba College of Engineering and Management Maharashtra Nagpur India Department of Computer Science and Engineering National Institute of Technology Chattisgarh Raipur India

Sarcasm is a type of communication that involves using words with meanings opposite to their literal definition to create humor or mock someone. This form of communication can be confusing as it often uses positive words to express negative feelings, making it difficult for people to understand the speaker’s intention. Detecting sarcasm in text can be challenging as it changes the polarity of the sentence and the difference between the words used and the way it is spoken. As a result, sarcasm detection in news, comments, or tweets on social media poses a challenge for researchers. In this research, various word-level features have been studied to detect sarcasm from three benchmark datasets, which include the creation of an N-gram probability dictionary, negation words, and PoS tags. Different machine learning and hybrid deep learning models have been examined and compared with handcrafted features and also with word embedding features. The results demonstrate a maximum accuracy of 87% using manual features with deep learning models and up to 92% accuracy using RF classifiers. © 2024, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd.

关键词： CNN LSTM N-gram probability Negation feature Word embedding

来源：评论

学校读者我要写书评

暂无评论

Analyzing Cross-Layer Design Approaches for Quality of Service in Wireless Networks 15

Analyzing Cross-Layer Design Approaches for Quality of Servi...

引用

15th International Conference on Computing Communication and Networking Technologies, ICCCNT 2024

作者： Shreenidhi, H.S. Acharjya, Kalyan Saraswat, Nidhi Jain School of Engineering and Technology Department of Computer Science and Engineering Karnataka Bangalore India Maharishi University of Information Technology Maharishi School of Engineering & technology Uttar Pradesh India Sanskriti University Department of Computer Science Engineering Uttar Pradesh Mathura India

ISBN: (纸本)9798350370249

Analyzing move-Layer layout procedures for satisfactory of provider (Qu 's) in Wi-Fi networks is of widespread importance to enhance network overall performance. Go-Layer optimization integrates distinct layers consisting of the physical, medium get right of entry to control (MAC), and community layers. Those layers are combined to successfully model the stop-to-cease network performance for each Wi-Fi and wired systems. Go-layer optimization can be used to improve the Qu's via numerous mechanisms, which include aid allocation, scheduling, adaptive modulation, and energy manipulate. These optimization techniques can be used to implement several Qu's-oriented functions together with safety in opposition to congestion, resilience to node and link failures, and adaptive visitors engineering. In this paper, we analyses numerous procedures to move-layer optimization that are appropriate for adapting the Qu's to dynamic situations in Wi-Fi networks. We talk the advantages and downsides of each method, and offer a complete overview on the modern-day studies trends. Eventually, we advise a framework for further research in this region. © 2024 IEEE.

关键词： Adaptive Optimization Performance Procedures Satisfactory Successfully Techniques

来源：评论

学校读者我要写书评

暂无评论

Intelligent Deep Learning Based Cybersecurity Phishing Email Detection and Classification

引用

computers, Materials & Continua 2023年第3期74卷 5901-5914页

作者： R.Brindha S.Nandagopal H.Azath V.Sathana Gyanendra Prasad Joshi Sung Won Kim Department of Computing Technologies SRM Institute of Science and TechnologyKattankulathur603203India Department of Computing Science and Engineering Nandha College of TechnologyErode638052India School of Computing Science and Engineering VIT Bhopal UniversityBhopal466114India Department of Computer Science and Engineering K.Ramakrishnan College of EngineeringTiruchirappalli621112India Department of Computer Science and Engineering Sejong UniversitySeoul05006Korea Department of Information and Communication Engineering Yeungnam UniversityGyeongsan-si38541Gyeongbuk-doKorea

Phishing is a type of cybercrime in which cyber-attackers pose themselves as authorized persons or entities and hack the victims’sensitive data.E-mails,instant messages and phone calls are some of the common modes used in *** the security models are continuously upgraded to prevent cyberattacks,hackers find innovative ways to target the *** this background,there is a drastic increase observed in the number of phishing emails sent to potential *** scenario necessitates the importance of designing an effective classification *** numerous conventional models are available in the literature for proficient classification of phishing emails,the Machine Learning(ML)techniques and the Deep Learning(DL)models have been employed in the *** current study presents an Intelligent Cuckoo Search(CS)Optimization Algorithm with a Deep Learning-based Phishing Email Detection and Classification(ICSOA-DLPEC)*** aim of the proposed ICSOA-DLPEC model is to effectually distinguish the emails as either legitimate or phishing *** the initial stage,the pre-processing is performed through three stages such as email cleaning,tokenization and stop-word ***,the N-gram approach is;moreover,the CS algorithm is applied to extract the useful feature ***,the CS algorithm is employed with the Gated Recurrent Unit(GRU)model to detect and classify phishing ***,the CS algorithm is used to fine-tune the parameters involved in the GRU *** performance of the proposed ICSOA-DLPEC model was experimentally validated using a benchmark dataset,and the results were assessed under several *** comparative studies were conducted,and the results confirmed the superior performance of the proposed ICSOA-DLPEC model over other existing *** proposed model achieved a maximum accuracy of 99.72%.

关键词： Phishing email data classification natural language processing deep learning cybersecurity

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：