检索结果-内蒙古大学图书馆

Feature-Based Augmentation in Sarcasm Detection Using Reverse Generative Adversarial Network

computers, Materials & Continua 2023年第12期77卷 3637-3657页

作者： Derwin Suhartono Alif Tri Handoyo Franz Adeta Junior Computer Science Department School of Computer ScienceBina Nusantara UniversityJakarta11480Indonesia Cyber Security Program Computer Science DepartmentSchool of Computer ScienceBina Nusantara UniversityJakarta11480Indonesia

Sarcasm detection in text data is an increasingly vital area of research due to the prevalence of sarcastic content in online *** study addresses challenges associated with small datasets and class imbalances in sarcasm detection by employing comprehensive data pre-processing and Generative Adversial Network(GAN)based augmentation on diverse datasets,including iSarcasm,SemEval-18,and *** research offers a novel pipeline for augmenting sarcasm data with Reverse Generative Adversarial Network(RGAN).The proposed RGAN method works by inverting labels between original and synthetic data during the training *** inversion of labels provides feedback to the generator for generating high-quality data closely resembling the original ***,the proposed RGAN model exhibits performance on par with standard GAN,showcasing its robust efficacy in augmenting text *** exploration of various datasets highlights the nuanced impact of augmentation on model performance,with cautionary insights into maintaining a delicate balance between synthetic and original *** methodological framework encompasses comprehensive data pre-processing and GAN-based augmentation,with a meticulous comparison against Natural Language Processing Augmentation(NLPAug)as an alternative augmentation ***,the F1-score of our proposed technique outperforms that of the synonym replacement augmentation technique using *** increase in F1-score in experiments using RGAN ranged from 0.066%to 1.054%,and the use of standard GAN resulted in a 2.88%increase in *** proposed RGAN model outperformed the NLPAug method and demonstrated comparable performance to standard GAN,emphasizing its efficacy in text data augmentation.

关键词： Data augmentation Generative Adversarial Network(GAN) Reverse GAN(RGAN) sarcasm detection

来源：评论

学校读者我要写书评

暂无评论

A Privacy-Preserving Data Aggregation Protocol for Internet of Vehicles with Federated Learning

IEEE Transactions on Intelligent Vehicles

引用

IEEE Transactions on Intelligent Vehicles 2024年 1-11页

作者： Xu, Zisang Zhang, Ruirui Liang, Wei Li, Kuan-Ching Gu, Ke Li, Xiong Huang, Jialun Computer and Communication Engineer Institute Changsha University of Science and Technology Changsha China School of Computer Science and Engineering Hunan University of Science and Technology Xiangtan China Department of Computer Science and Information Engineering Providence University Taichung Taiwan Institute for Cyber Security School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu China

Federated learning (FL) is widely used in various fields because it can guarantee the privacy of the original data source. However, in data-sensitive fields such as Internet of Vehicles (IoV), insecure communication channels, semi-trusted RoadSide Unit (RSU), and collusion between vehicles and the RSU may lead to leakage of model parameters. Moreover, when aggregating data, since different vehicles usually have different computing resources, vehicles with relatively insufficient computing resources will affect the data aggregation efficiency. Therefore, in order to solve the privacy leakage problem and improve the data aggregation efficiency, this paper proposes a privacy-preserving data aggregation protocol for IoV with FL. Firstly, the protocol is designed based on methods such as shamir secret sharing scheme, pallier homomorphic encryption scheme and blinding factor protection, which can guarantee the privacy of model parameters. Secondly, the protocol improves the data aggregation efficiency by setting dynamic training time windows. Thirdly, the protocol reduces the frequent participations of Trusted Authority (TA) by optimizing the fault-tolerance mechanism. Finally, the security analysis proves that the proposed protocol is secure, and the performance analysis results also show that the proposed protocol has high computation and communication efficiency. IEEE

关键词： Vehicles

来源：评论

学校读者我要写书评

暂无评论

LCEFL: A Lightweight Contribution Evaluation Approach for Federated Learning

引用

IEEE Transactions on Mobile Computing 2025年第7期24卷 6643-6657页

作者： Guo, Jingjing Li, Jiaxing Liu, Zhiquan Xiong, Yupeng Ma, Yong Vasilakos, Athanasios V. Li, Xinghua Ma, Jianfeng Xidian University School of Cyber Engineering Xi’an710071 China Jinan University College of Cyber Security Guangzhou510632 China Jiangxi Normal University Department of Computer Science and Technology Nanchang330022 China Department of Networks and Communications College of Computer Science and Information Technology Dammam31441 Saudi Arabia Grimstad4879 Norway

The prerequisite for implementing incentive mechanisms and reliable participant selection schemes in federated learning is to obtain the contribution of each participant. Available evaluation methods for participant contributions require the server to possess a test dataset, often impractical. Additionally, the excessively high complexity of these works is unacceptable when training complex models in large-scale federated learning system. To address these issues, we propose a lightweight contribution evaluation method for federated learning participants, named LCEFL, based on model projection theory, which does not require the server to provide a test dataset. In addition, a model compression method is designed to be used in LCEFL to reduce the computational complexity. Furthermore, a trusted aggregation method based on LCEFL is proposed, where the weight of each participant's local model is determined by its trust level, which can be calculated using its contribution evaluation result. Experimental results show that LCEFL can achieve nearly the same accuracy as schemes based on Shapley Value, while significantly reducing computational overhead by more than 50%. Compared to available aggregation methods, the proposed trusted aggregation scheme is able to accelerate the convergence speed of the global model and improve its accuracy by 2% to 45%. © 2002-2012 IEEE.

关键词： Federated Learning Computational Modeling Accuracy Servers Training Data Models Machine Learning Mobile Computing Particle Measurements Convergence Federated Learning Contribution Evaluation Trust Evaluation Model Projection Theory Secure Aggregation Federated Learning Computational Complexity Test Dataset Global Model Selection Of Participants Model Projections Aggregation Method Shapley Value Speed Of The Model Contributions Of Participants Number Of Participants Accuracy Of Model Computational Cost Local Data Learning Task Layer Model Participation Outcomes Updated Model Value Of Participation Model Aggregation Server Aggregates Global Convergence Iterative Rounds Trust Value Local Updates MNIST Dataset Total Amount Of Data Local Training Aggregation Algorithm Complex Model Structure

来源：评论

学校读者我要写书评

暂无评论

ArchSentry: Enhanced Android Malware Detection via Hierarchical Semantic Extraction

引用

IEEE Transactions on Network and Service Management 2025年第3期22卷 2822-2837页

作者： Wang, Tianbo Liu, Mengyao Li, Huacheng Zhao, Lei Jiang, Changnan Xia, Chunhe Cui, Baojiang Beihang University School of Cyber Science and Technology Beijing100191 China Beihang University School of Computer Science and Engineering Beijing China Wuhan University School of Cyber Science and Engineering Wuhan China Beijing University of Posts and Telecommunications School of Cyberspace Security Beijing China

Android malware poses a significant challenge for mobile platforms. To evade detection, contemporary malware variants use API substitution or obfuscation techniques to hide malicious activities and mask their shallow semantic characteristics. However, existing research lacks analysis of the hierarchical semantic associated with Android apps. To address this problem, we propose ArchSentry, an enhanced Android malware detection via hierarchical semantic extraction. First, we select entities and their relationships relevant to Android software behavior through the software architecture and represent them using a heterogeneous graph. Then, we structure meta-paths to represent rich semantic information to achieve semantic enhancement and improve efficiency. Next, we design a meta-path semantic selection method based on KL Divergence to identify and eliminate redundant features. To achieve a comprehensive representation of the overall software semantics and improve performance, we construct a feature fusion approach based on Restricted Boltzmann Machines (RBM) and AutoEncoder (AE) during the pre-training phase, while preserving the probability distribution characteristics of various meta-paths. Finally, Deep Neural Networks (DNN) process fusion features for comprehensive feature sets. Experimental results on real-world application samples indicate that ArchSentry achieves a remarkable 99.2% detection rate for Android malware, with a low false positive rate below 1%. These results surpass the performance of current state-of-the-art approaches. © 2004-2012 IEEE.

关键词： Android malware

来源：评论

学校读者我要写书评

暂无评论

A Post-Quantum Privacy-Preserving Payment Protocol in Vehicle to Grid Networks

IEEE Transactions on Intelligent Vehicles

引用

IEEE Transactions on Intelligent Vehicles 2024年 1-13页

作者： Chen, Ying He, Debiao Bao, Zijian Luo, Min Choo, Kim-Kwang Raymond School of Cyber Science and Engineering Wuhan University Wuhan China Department of Information Systems and Cyber Security University of Texas at San Antonio San Antonio TX USA

In Vehicle to Grid (V2G) networks, the frequent charging and discharging activities of electric vehicles at specific locations underscore the importance to address the challenges associated with identity and location privacy leakages. Existing solutions are mostly built on difficult problems in number theory, which are known to be insecure against attacks carried out using quantum computers. In this paper, we first propose Azalea, a logarithmic lattice-based revocable and linkable ring signature scheme, and use it as a building block and transform it to a postquantum secure revocable privacy-preserving payment protocol for V2G networks. The latter protocol's security relies on the assumption of Module Learning with Errors (M-LWE) problem. We also give the security proof of AZALEA in the random oracle model (ROM), and the security analysis indicates that our payment protocol meets the privacy demands. Furthermore, the efficiency evaluation results demonstrate the potential utility of our construction in V2G network settings. IEEE

关键词： Authentication

来源：评论

学校读者我要写书评

暂无评论

Malicious PE File Detection Using Machine Learning: An Analysis of Header Features

Malicious PE File Detection Using Machine Learning: An Analy...

引用

2024 IEEE International Conference on Computing, Semiconductor, Mechatronics, Intelligent Systems and Communications, COSMIC 2024

作者： Shyam Kumar, S. Shetty, Jyothi Department of Cyber Security Nitte574 110 India Department of Computer Science and Engineering Nitte574 110 India

ISBN: (纸本)9798331517892

Portable Executable (PE) files serve as the foundation of the Windows operating system, containing executable code, object code, and Dynamic Link Libraries (DLLs). Their importance lies in their capacity to execute tasks and run programs within the Windows environment, rendering them essential to software execution. Comprehension of PE files is crucial in cybersecurity due to their prevalence and potential for harboring malicious software. Analysis of PE files enables researchers and analysts to uncover insights into their structure, behavior, and dependencies, which is essential for detecting and categorizing potential threats. PE file analysis provides critical information about a program's characteristics, facilitating the identification of indicators of compromise and potential security risks. Through examination of PE headers, sections, and attributes, analysts can gain insights into the file's functionality, dependencies, and potential security vulnerabilities. This knowledge equips cybersecurity professionals to develop robust detection and mitigation strategies, safeguarding systems and networks against cyber threats. In this paper, we examine the intricate details of PE files, utilizing tools such as PE-bear, PEinfo, CFF Explorer, and PEStudio to conduct comprehensive analysis. Our study employs machine learning techniques, specifically the Random Forest algorithm, to enhance malware detection capabilities. By preprocessing and augmenting the dataset with new executable files, we aim to enhance the efficacy of our analysis and provide valuable insights for cybersecurity research and defense strategies. © 2024 IEEE.

关键词： cybersecurity

来源：评论

学校读者我要写书评

暂无评论

Advancing Android Malware Detection with BioSentinel Neural Network using Hybrid Deep Learning Techniques 18

Advancing Android Malware Detection with BioSentinel Neural ...

引用

18th INDIAcom;11th International Conference on Computing for Sustainable Global Development, INDIACom 2024

作者： Rani, D. Sandhya Gnaneshwar, K. Sampurnima Pattem, K. Sekhar, Soma Krishna, G. Bala Kakarla, Shirisha Cvr College of Engineering Department of Computer Science Engineering Rangareddy Ibrahimpatnam India Sreenidhi Institute of Science and Technology Department of Computer Science Engineering-Cyber Security Hyderabad India Vardhaman College of Engineering Department of Computer Science Engineering Hyderabad India

ISBN: (纸本)9789380544519

The paper introduces the BioSentinel Neural Network (BSNN), a novel hybrid deep learning model designed to enhance malware detection, particularly focusing on zero-day threats. The BSNN model integrates diverse neural network architectures, leveraging Graph Neural Networks (GNNs) for feature extraction and Transformers for sequential data analysis. It demonstrates significant improvements in accuracy (93.16%), precision (90.89%), recall (88.19%), and F1-score (90.45%) over traditional methods. The model exhibits a high detection rate of 88% for zero-day malware from 1000 samples tested and shows computational efficiency with an average inference time of 30 milliseconds. This research addresses critical gaps in current malware detection approaches, highlighting the potential of hybrid deep learning techniques in the cybersecurity domain. © 2024 Bharati Vidyapeeth, New Delhi.

关键词： cybersecurity Deep Learning Hybrid Neural Networks Malware Detection Real-Time Analysis Zero-Day Threats

来源：评论

学校读者我要写书评

暂无评论

Energy Efficiency Maximization for Cooperative NOMA with Hardware Impairments

引用

China Communications 2024年第12期21卷 80-91页

作者： Wang Zhengqiang Chang Ruifei Wan Xiaoyu Fan Zifu Duo Bin School of Communication and Information Engineering Chongqing University of Posts and TelecommunicationsChongqing 400065China College of Computer Science and Cyber Security Chengdu University of TechnologyChengdu 610059China

The massive connectivity and limited energy pose significant challenges to deploy the enormous devices in energy-efficient and environmentally friendly in the Internet of Things(IoT).Motivated by these challenges,this paper investigates the energy efficiency(EE)maximization problem for downlink cooperative non-orthogonal multiple access(C-NOMA)systems with hardware impairments(HIs).The base station(BS)communicates with several users via a half-duplex(HD)amplified-and-forward(AF)***,we formulate the EE maximization problem of the system under HIs by jointly optimizing transmit power and power allocated coefficient(PAC)at BS,and transmit power at the *** original EE maximization problem is a non-convex problem,which is challenging to give the optimal solution ***,we use fractional programming to convert the EE maximization problem as a series of subtraction form ***,variable substitution and block coordinate descent(BCD)method are used to handle the ***,a resource allocation algorithm is proposed to maximize the EE of the ***,simulation results show that the proposed algorithm outperforms the downlink cooperative orthogonal multiple access(C-OMA)scheme.

关键词： block coordinate descent cooperative non-orthogonal multiple access energy efficiency hardware impairments resource allocation

来源：评论

学校读者我要写书评

暂无评论

EEG-VARNet: An Advanced Deep Learning Model for AccuratePrediction and Classification of Major Depressive Disorder Using EEG Data 6

EEG-VARNet: An Advanced Deep Learning Model for AccuratePred...

引用

6th International Conference on cybernetics, Cognition and Machine Learning Applications, ICCCMLA 2024

作者： Mahender, Udutala Sathiyasuntharam, V. Annamalai University Department of Computer Science and Engineering Tamilnadu Chidambaram India CMR Engineering College Department of Cyber Security Hyderabad India

ISBN: (纸本)9798331505790

The correct and early identification of Major Depressive Disorder (MDD) is challenging because of its complicated causes and symptoms that are similar to those of other mental diseases. This study introduces a new method for forecasting and categorizing MDD by utilizing EEG data, with a specific emphasis on Valence and Arousal measurements. Our approach combines sophisticated methods for extracting and selecting features with deeplearning models. Firstly, EEG signals are subjected to preprocessing in order to eliminate artifacts and standardize the quality of the data. The Boruta algorithm is utilized to extract features, which are then selected using the improved Cuckoo Search Algorithm. The suggested model, referred to as EEG-VARNet, utilizes two CNN models (Xception V4) and an SVM classifier to capture both spatial and temporal dependencies present in the EEG data. The model undergoes hyperparameter adjustment to maximize performance. The results of our study show substantial enhancements in the prediction of MDD with an accuracy rate of 9 8. 2 %, a precision rate of 96.43 %, and a recall rate of 9 7. 8 9 %. The measurements demonstrate the model's exceptional ability to diagnose MDD, emphasizing its promise as a dependable tool for clinical use. © 2024 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Effective Tool Augmented Multi-Agent Framework for Data Analysis

引用

Data Intelligence 2024年第4期6卷 923-945页

作者： Xilin Zhang Zhixin Mao Ziwen Chen Shen Gao School of Automation Engineering University of Electronic Science and Technology of ChinaChengdu 611731China School of Computer Science and Engineering(School of Cyber Security) University of Electronic Science and Technology of ChinaChengdu 611731China

Data analysis tasks aim to provide insightful analysis for given data by incorporating background knowledge of the represented phenomenon, which in turn supports decision-making. While existing large language models(LLMs) can describe data trends, they still lag behind human data analysts in terms of integrating external knowledge and in-depth data analysis. Therefore, we propose a multi-agent data analysis framework based on LLMs. The framework decomposes the data analysis task into subtasks by employing three different agents. By empowering agents with the ability to utilize data search tools, the framework enables them to search for arbitrary relevant knowledge during the analysis process, leading to more insightful analysis. Moreover, to enhance the quality of the analysis results, we propose a multi-stage iterative optimization method that iteratively performs data analysis to form more in-depth conclusions. To validate the performance of our framework, we apply it to three real-world problems in the research development of higher education in China data. Experimental results demonstrate that our approach can achieve more insightful data analysis results compared to directly using LLMs alone.

关键词： Exploratory data analysis Multi-agent Large language models

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：