期刊文献+
共找到139篇文章
< 1 2 7 >
每页显示 20 50 100
IDS-INT:Intrusion detection system using transformer-based transfer learning for imbalanced network traffic 被引量:3
1
作者 Farhan Ullah Shamsher Ullah +1 位作者 Gautam Srivastava Jerry Chun-Wei Lin 《Digital Communications and Networks》 SCIE CSCD 2024年第1期190-204,共15页
A network intrusion detection system is critical for cyber security against llegitimate attacks.In terms of feature perspectives,network traffic may include a variety of elements such as attack reference,attack type,a... A network intrusion detection system is critical for cyber security against llegitimate attacks.In terms of feature perspectives,network traffic may include a variety of elements such as attack reference,attack type,a subcategory of attack,host information,malicious scripts,etc.In terms of network perspectives,network traffic may contain an imbalanced number of harmful attacks when compared to normal traffic.It is challenging to identify a specific attack due to complex features and data imbalance issues.To address these issues,this paper proposes an Intrusion Detection System using transformer-based transfer learning for Imbalanced Network Traffic(IDS-INT).IDS-INT uses transformer-based transfer learning to learn feature interactions in both network feature representation and imbalanced data.First,detailed information about each type of attack is gathered from network interaction descriptions,which include network nodes,attack type,reference,host information,etc.Second,the transformer-based transfer learning approach is developed to learn detailed feature representation using their semantic anchors.Third,the Synthetic Minority Oversampling Technique(SMOTE)is implemented to balance abnormal traffic and detect minority attacks.Fourth,the Convolution Neural Network(CNN)model is designed to extract deep features from the balanced network traffic.Finally,the hybrid approach of the CNN-Long Short-Term Memory(CNN-LSTM)model is developed to detect different types of attacks from the deep features.Detailed experiments are conducted to test the proposed approach using three standard datasets,i.e.,UNsWNB15,CIC-IDS2017,and NSL-KDD.An explainable AI approach is implemented to interpret the proposed method and develop a trustable model. 展开更多
关键词 Network intrusion detection Transfer learning Features extraction imbalance data Explainable AI CYBERSECURITY
下载PDF
Handling Class Imbalance in Online Transaction Fraud Detection
2
作者 Kanika Jimmy Singla +3 位作者 Ali Kashif Bashir Yunyoung Nam Najam UI Hasan Usman Tariq 《Computers, Materials & Continua》 SCIE EI 2022年第2期2861-2877,共17页
With the rise of internet facilities,a greater number of people have started doing online transactions at an exponential rate in recent years as the online transaction system has eliminated the need of going to the ba... With the rise of internet facilities,a greater number of people have started doing online transactions at an exponential rate in recent years as the online transaction system has eliminated the need of going to the bank physically for every transaction.However,the fraud cases have also increased causing the loss of money to the consumers.Hence,an effective fraud detection system is the need of the hour which can detect fraudulent transactions automatically in real-time.Generally,the genuine transactions are large in number than the fraudulent transactions which leads to the class imbalance problem.In this research work,an online transaction fraud detection system using deep learning has been proposed which can handle class imbalance problem by applying algorithm-level methods which modify the learning of the model to focus more on the minority class i.e.,fraud transactions.A novel loss function named Weighted Hard-Reduced Focal Loss(WH-RFL)has been proposed which has achieved maximum fraud detection rate i.e.,True PositiveRate(TPR)at the cost of misclassification of few genuine transactions as high TPR is preferred over a high True Negative Rate(TNR)in fraud detection system and same has been demonstrated using three publicly available imbalanced transactional datasets.Also,Thresholding has been applied to optimize the decision threshold using cross-validation to detect maximum number of frauds and it has been demonstrated by the experimental results that the selection of the right thresholding method with deep learning yields better results. 展开更多
关键词 Class imbalance deep learning fraud detection loss function THRESHOLDING
下载PDF
Dealing with the Class Imbalance Problem in the Detection of Fake Job Descriptions
3
作者 Minh Thanh Vo Anh H.Vo +2 位作者 Trang Nguyen Rohit Sharma Tuong Le 《Computers, Materials & Continua》 SCIE EI 2021年第7期521-535,共15页
In recent years,the detection of fake job descriptions has become increasingly necessary because social networking has changed the way people access burgeoning information in the internet age.Identifying fraud in job ... In recent years,the detection of fake job descriptions has become increasingly necessary because social networking has changed the way people access burgeoning information in the internet age.Identifying fraud in job descriptions can help jobseekers to avoid many of the risks of job hunting.However,the problem of detecting fake job descriptions comes up against the problem of class imbalance when the number of genuine jobs exceeds the number of fake jobs.This causes a reduction in the predictability and performance of traditional machine learning models.We therefore present an efficient framework that uses an oversampling technique called FJD-OT(Fake Job Description Detection Using Oversampling Techniques)to improve the predictability of detecting fake job descriptions.In the proposed framework,we apply several techniques including the removal of stop words and the use of a tokenizer to preprocess the text data in the first module.We then use a bag of words in combination with the term frequency-inverse document frequency(TF-IDF)approach to extract the features from the text data to create the feature dataset in the second module.Next,our framework applies k-fold cross-validation,a commonly used technique to test the effectiveness of machine learning models,that splits the experimental dataset[the Employment Scam Aegean(ESA)dataset in our study]into training and test sets for evaluation.The training set is passed through the third module,an oversampling module in which the SVMSMOTE method is used to balance data before training the classifiers in the last module.The experimental results indicate that the proposed approach significantly improves the predictability of fake job description detection on the ESA dataset based on several popular performance metrics. 展开更多
关键词 Fake job description detection class imbalance problem oversampling techniques
下载PDF
Detection of chromosomal imbalance in oligodendroglial tumors by comparative genomic hybridization
4
作者 孙翠云 《外科研究与新技术》 2011年第3期216-216,共1页
Objective To investigate the relationship between genomic DNA imbalance in oligodendroglial tumors and its different classification. Methods 16 oligodendrogliomas and 17 anaplastic oligodendrogliomas were investigated... Objective To investigate the relationship between genomic DNA imbalance in oligodendroglial tumors and its different classification. Methods 16 oligodendrogliomas and 17 anaplastic oligodendrogliomas were investigated by comparative genomic hybridization on Paraffin-Embedded tissue samples,and the chromosomal genomic DNA imbalances were analyzed. Results Chromosome DNA imbalance rates in oligodendrogliomas 展开更多
关键词 detection of chromosomal imbalance in oligodendroglial tumors by comparative genomic hybridization
下载PDF
MEM-TET: Improved Triplet Network for Intrusion Detection System 被引量:2
5
作者 Weifei Wang Jinguo Li +1 位作者 Na Zhao Min Liu 《Computers, Materials & Continua》 SCIE EI 2023年第7期471-487,共17页
With the advancement of network communication technology,network traffic shows explosive growth.Consequently,network attacks occur frequently.Network intrusion detection systems are still the primary means of detectin... With the advancement of network communication technology,network traffic shows explosive growth.Consequently,network attacks occur frequently.Network intrusion detection systems are still the primary means of detecting attacks.However,two challenges continue to stymie the development of a viable network intrusion detection system:imbalanced training data and new undiscovered attacks.Therefore,this study proposes a unique deep learning-based intrusion detection method.We use two independent in-memory autoencoders trained on regular network traffic and attacks to capture the dynamic relationship between traffic features in the presence of unbalanced training data.Then the original data is fed into the triplet network by forming a triplet with the data reconstructed from the two encoders to train.Finally,the distance relationship between the triples determines whether the traffic is an attack.In addition,to improve the accuracy of detecting unknown attacks,this research proposes an improved triplet loss function that is used to pull the distances of the same class closer while pushing the distances belonging to different classes farther in the learned feature space.The proposed approach’s effectiveness,stability,and significance are evaluated against advanced models on the Android Adware and General Malware Dataset(AAGM17),Knowledge Discovery and Data Mining Cup 1999(KDDCUP99),Canadian Institute for Cybersecurity Group’s Intrusion Detection Evaluation Dataset(CICIDS2017),UNSW-NB15,Network Security Lab-Knowledge Discovery and Data Mining(NSL-KDD)datasets.The achieved results confirmed the superiority of the proposed method for the task of network intrusion detection. 展开更多
关键词 Intrusion detection memory-augmented autoencoder deep metric learning imbalance data
下载PDF
Enhanced Coyote Optimization with Deep Learning Based Cloud-Intrusion Detection System 被引量:1
6
作者 Abdullah M.Basahel Mohammad Yamin +1 位作者 Sulafah M.Basahel E.Laxmi Lydia 《Computers, Materials & Continua》 SCIE EI 2023年第2期4319-4336,共18页
Cloud Computing(CC)is the preference of all information technology(IT)organizations as it offers pay-per-use based and flexible services to its users.But the privacy and security become the main hindrances in its achi... Cloud Computing(CC)is the preference of all information technology(IT)organizations as it offers pay-per-use based and flexible services to its users.But the privacy and security become the main hindrances in its achievement due to distributed and open architecture that is prone to intruders.Intrusion Detection System(IDS)refers to one of the commonly utilized system for detecting attacks on cloud.IDS proves to be an effective and promising technique,that identifies malicious activities and known threats by observing traffic data in computers,and warnings are given when such threatswere identified.The current mainstream IDS are assisted with machine learning(ML)but have issues of low detection rates and demanded wide feature engineering.This article devises an Enhanced Coyote Optimization with Deep Learning based Intrusion Detection System for Cloud Security(ECODL-IDSCS)model.The ECODL-IDSCS model initially addresses the class imbalance data problem by the use of Adaptive Synthetic(ADASYN)technique.For detecting and classification of intrusions,long short term memory(LSTM)model is exploited.In addition,ECO algorithm is derived to optimally fine tune the hyperparameters related to the LSTM model to enhance its detection efficiency in the cloud environment.Once the presented ECODL-IDSCS model is tested on benchmark dataset,the experimental results show the promising performance of the ECODL-IDSCS model over the existing IDS models. 展开更多
关键词 Intrusion detection system cloud security coyote optimization algorithm class imbalance data deep learning
下载PDF
Improving the Detection Rate of Rarely Appearing Intrusions in Network-Based Intrusion Detection Systems
7
作者 Eunmok Yang Gyanendra Prasad Joshi Changho Seo 《Computers, Materials & Continua》 SCIE EI 2021年第2期1647-1663,共17页
In network-based intrusion detection practices,there are more regular instances than intrusion instances.Because there is always a statistical imbalance in the instances,it is difficult to train the intrusion detectio... In network-based intrusion detection practices,there are more regular instances than intrusion instances.Because there is always a statistical imbalance in the instances,it is difficult to train the intrusion detection system effectively.In this work,we compare intrusion detection performance by increasing the rarely appearing instances rather than by eliminating the frequently appearing duplicate instances.Our technique mitigates the statistical imbalance in these instances.We also carried out an experiment on the training model by increasing the instances,thereby increasing the attack instances step by step up to 13 levels.The experiments included not only known attacks,but also unknown new intrusions.The results are compared with the existing studies from the literature,and show an improvement in accuracy,sensitivity,and specificity over previous studies.The detection rates for the remote-to-user(R2L)and user-to-root(U2L)categories are improved significantly by adding fewer instances.The detection of many intrusions is increased from a very low to a very high detection rate.The detection of newer attacks that had not been used in training improved from 9%to 12%.This study has practical applications in network administration to protect from known and unknown attacks.If network administrators are running out of instances for some attacks,they can increase the number of instances with rarely appearing instances,thereby improving the detection of both known and unknown new attacks. 展开更多
关键词 Intrusion detection statistical imbalance SMO machine learning network security
下载PDF
基于端口注意力与通道空间注意力的网络异常流量检测 被引量:2
8
作者 肖斌 甘昀 +2 位作者 汪敏 张兴鹏 王照星 《计算机应用》 CSCD 北大核心 2024年第4期1027-1034,共8页
网络异常流量检测是网络安全保护重要组成部分之一。目前,基于深度学习的异常流量检测方法都是将端口号属性与其他流量属性同等对待,忽略了端口号的重要性。为了提高异常流量检测性能,借鉴注意力思想,提出一个卷积神经网络(CNN)结合端... 网络异常流量检测是网络安全保护重要组成部分之一。目前,基于深度学习的异常流量检测方法都是将端口号属性与其他流量属性同等对待,忽略了端口号的重要性。为了提高异常流量检测性能,借鉴注意力思想,提出一个卷积神经网络(CNN)结合端口注意力模块(PAM)和通道空间注意力模块(CBAM)的网络异常流量检测模型。首先,将原始网络流量作为PAM的输入,分离得到端口号属性送入全连接层,得到学习后的端口注意力权重值,并与其他流量属性点乘,输出端口注意力后的流量数据;其次,将流量数据转换成灰度图,利用CNN和CBAM更充分地提取特征图在通道和空间上的信息;最后,使用焦点损失函数解决数据不平衡的问题。所提PAM具有参数量少、即插即用和普遍适用的优点。在CICIDS2017数据集上,所提模型的异常流量检测二分类任务准确率为99.18%,多分类任务准确率为99.07%,对只有少数训练样本的类别也有较高的识别率。 展开更多
关键词 异常流量检测 注意力机制 数据不平衡 轻量级网络 通道空间注意力模块
下载PDF
基于MobileViT的轻量型入侵检测模型研究
9
作者 姚军 孙方超 《现代电子技术》 北大核心 2024年第19期33-39,共7页
为解决入侵检测中数据不平衡对神经网络模型训练的影响和模型参数量高的问题,提出一种基于改进MobileViT的入侵检测模型。首先,使用方差分析提取对检测结果影响较高的特征,将提取后的特征转化为图像型数据,将其输入至MobileViT网络;其次... 为解决入侵检测中数据不平衡对神经网络模型训练的影响和模型参数量高的问题,提出一种基于改进MobileViT的入侵检测模型。首先,使用方差分析提取对检测结果影响较高的特征,将提取后的特征转化为图像型数据,将其输入至MobileViT网络;其次,针对占比较少的攻击流量,采用焦点损失函数自适应地调整攻击流量的损失贡献,使模型更加专注于不平衡的攻击流量;最后,为解决神经元死亡问题,使用GeLU激活函数替换MobileViT网络中MV2的ReLU6激活函数,加快模型收敛速度。实验结果表明,改进的MobileViT模型参数量仅为5.67 MB,与Shufflenet、Mobilenet相比拥有最少的参数量,模型的准确率、召回率以及F_(1)分数分别达到了98.40%、96.49%、95.17%。 展开更多
关键词 入侵检测 焦点损失函数 数据不平衡 MobileViT GeLU 方差分析
下载PDF
结合遗传算法的RF-DBN入侵检测方法
10
作者 任俊玲 诸于铭 《中国科技论文》 CAS 2024年第8期937-944,共8页
针对目前不平衡数据集少数类攻击样本识别率较低的问题,提出一种BorderlineSMOTE、随机森林和遗传算法(genetic algorithm,GA)-深度信念网络(deep belief network,DBN)相结合的入侵检测方法。首先采用BorderlineSMOTE对少数类样本进行... 针对目前不平衡数据集少数类攻击样本识别率较低的问题,提出一种BorderlineSMOTE、随机森林和遗传算法(genetic algorithm,GA)-深度信念网络(deep belief network,DBN)相结合的入侵检测方法。首先采用BorderlineSMOTE对少数类样本进行过采样,减少数据集的不平衡度;然后使用随机森林算法实现正异常数据分类,筛选出异常数据;最后采用经GA优化的DBN网络对异常数据进行进一步分类。使用网络安全数据集CICIDS2017进行验证,该方法的准确率达到了99.85%,而且少数类样本的识别精度也有明显提高。 展开更多
关键词 随机森林 遗传算法 BorderlineSMOTE 深度信念网络 数据不平衡 入侵检测
下载PDF
融合1D-CNN与BiGRU的类不平衡流量异常检测
11
作者 陈虹 齐兵 +2 位作者 金海波 武聪 张立昂 《计算机应用》 CSCD 北大核心 2024年第8期2493-2499,共7页
网络流量异常检测是利用各种检测技术分析判断网络流量,发现网络中潜在的攻击,是一种有效的网络安全防护方法。针对高维海量数据和不同攻击类别的网络流量数据不均衡而导致检测准确率低、误报率高的问题,提出一种融合一维卷积神经网络(1... 网络流量异常检测是利用各种检测技术分析判断网络流量,发现网络中潜在的攻击,是一种有效的网络安全防护方法。针对高维海量数据和不同攻击类别的网络流量数据不均衡而导致检测准确率低、误报率高的问题,提出一种融合一维卷积神经网络(1D-CNN)和双向门控循环单元(BiGRU)的类不平衡流量异常检测模型。首先,针对类不平衡数据,通过使用改进的合成少数类过采样技术(SMOTE)即Borderline-SMOTE和基于高斯混合模型(GMM)的欠采样聚类技术进行平衡处理;然后,使用1D-CNN提取数据的局部特征,并利用BiGRU更好地提取数据中的时序特征;最后,在UNSW-NB15数据集对所提模型进行验证,所提模型的准确率为98.12%,误报率为1.28%。结果表明,所提模型提高了对少数攻击的识别率,检测精度高于其他经典机器学习和深度学习模型。 展开更多
关键词 流量异常检测 不平衡处理 特征选择 卷积神经网络 双向门控循环单元
下载PDF
基于VAE-CWGAN和特征统计重要性融合的网络入侵检测方法 被引量:3
12
作者 刘涛涛 付钰 +1 位作者 王坤 段雪源 《通信学报》 EI CSCD 北大核心 2024年第2期54-67,共14页
针对传统入侵检测方法受限于数据集类不平衡以及所选特征代表性不强等问题,提出一种基于VAE-CWGAN和特征统计重要性融合的检测方法。首先,为提升数据质量对数据集进行预处理;其次,搭建VAE-CWGAN模型生成新样本以解决数据集类不平衡问题... 针对传统入侵检测方法受限于数据集类不平衡以及所选特征代表性不强等问题,提出一种基于VAE-CWGAN和特征统计重要性融合的检测方法。首先,为提升数据质量对数据集进行预处理;其次,搭建VAE-CWGAN模型生成新样本以解决数据集类不平衡问题,使分类模型不再偏向于多数类;再次,使用标准差、中值均值差对特征进行排序,并融合其统计重要性来进行特征选择旨在获得代表性更强的特征,从而使模型更好地学习数据信息;最后,通过一维卷积神经网络对特征选择后的混合数据集进行分类。实验结果表明,所提方法在NSL-KDD、UNSW-NB15和CIC-IDS-2017数据集上都表现出较好的性能优势,准确率分别为98.95%、96.24%和99.92%,有效提升了入侵检测性能。 展开更多
关键词 入侵检测 网络流量 类不平衡 特征选择 统计重要性融合
下载PDF
基于语义分割的侧扫声纳管线目标检测方法
13
作者 郑根 徐会希 +1 位作者 赵建虎 杨文林 《海洋测绘》 CSCD 北大核心 2024年第2期9-13,共5页
为提高侧扫声纳图像中管线目标检测的自动化程度及效率,提出了一种基于语义分割的水下管线目标检测方法。首先通过构建高效语义分割网络主干,提高网络计算速度并降低网络对计算机硬件性能的需求;其次给出了一种针对管线目标特点的加权... 为提高侧扫声纳图像中管线目标检测的自动化程度及效率,提出了一种基于语义分割的水下管线目标检测方法。首先通过构建高效语义分割网络主干,提高网络计算速度并降低网络对计算机硬件性能的需求;其次给出了一种针对管线目标特点的加权交叉熵损失函数,解决了因类间数量不均衡导致的网络训练困难问题。以多种复杂条件下侧扫声纳实测数据进行了水下管线检测试验,结果表明,该方法在取得和经典网络相近精度的情况下,速度提升了2.7倍,可达52.6FPS,实现了水下管线的快速、准确检测。 展开更多
关键词 水下目标检测 侧扫声纳图像 深度学习 语义分割 网络优化 类间不平衡
下载PDF
基于生成对抗网络的系统调用主机入侵检测技术
14
作者 樊燚 胡涛 伊鹏 《计算机科学》 CSCD 北大核心 2024年第10期408-415,共8页
程序的系统调用信息是检测主机异常的重要数据,然而异常发生的次数相对较少,这使得收集到的系统调用数据往往存在数据不均衡的问题。较少的异常系统调用数据使得检测模型无法充分理解程序的异常行为模式,导致入侵检测的准确率较低、误... 程序的系统调用信息是检测主机异常的重要数据,然而异常发生的次数相对较少,这使得收集到的系统调用数据往往存在数据不均衡的问题。较少的异常系统调用数据使得检测模型无法充分理解程序的异常行为模式,导致入侵检测的准确率较低、误报率较高。针对以上问题,提出了一种基于生成对抗网络的系统调用主机入侵检测方法,通过对异常系统调用数据的增强,缓解数据不平衡的问题。首先将程序的系统调用轨迹划分成固定长度的N-Gram序列,其次使用SeqGAN从异常数据的N-Gram序列中生成合成的N-Gram序列,生成的异常数据与原始数据集相结合,用于训练入侵检测模型。在一个主机系统调用数据集ADFA-LD及一个安卓系统调用数据集Drebin上进行了实验,所提方法的检测准确率分别为0.986和0.989,误报率分别为0.011和0,检测效果优于现有的基于混合神经网络的模型、WaveNet、Relaxed-SVM及RNN-VED的入侵检测研究方法。 展开更多
关键词 主机入侵检测 系统调用 生成对抗网络 深度学习 数据不均衡
下载PDF
一种基于半监督学习的网络异常流量检测方法
15
作者 钟昱 黄振南 +1 位作者 谢惠超 陈宁江 《广西大学学报(自然科学版)》 CAS 北大核心 2024年第3期563-574,共12页
针对网络流量数据存在标记样本获取困难、实际数据类别不平衡等问题,提出一种合成数据增强的半监督网络异常流量检测方法(SEASAND)。SEASAND利用无标记数据辅助模型学习,只需少量的有标签数据即可达到较高识别准确率,降低了训练成本。... 针对网络流量数据存在标记样本获取困难、实际数据类别不平衡等问题,提出一种合成数据增强的半监督网络异常流量检测方法(SEASAND)。SEASAND利用无标记数据辅助模型学习,只需少量的有标签数据即可达到较高识别准确率,降低了训练成本。考虑一致性正则和熵最小化原则,通过混合采样解决网络流量数据不平衡的问题,并采用混合样本算法对样本进行二次数据增强,提高了对无标记数据的利用效率。最后利用一维残差网络Resnet1D 18对数据增强后的数据集进行训练。SEASAND在KDDCup9910、UNSW-NB15、CICIDS2017数据集上进行仿真实验,结果表明,与相关算法对比,SEASAND在少样本、多分类问题上具有较好的性能,降低了对有标记样本量的需求。 展开更多
关键词 半监督学习 网络异常流量检测 混合采样 数据不平衡
下载PDF
Community-detection-based Approaches for Distribution Network Partition
16
作者 Mingming Mao Zaijun Wu +2 位作者 Dongliang Xu Junjun Xu Qinran Hu 《CSEE Journal of Power and Energy Systems》 SCIE EI CSCD 2024年第5期1965-1976,共12页
A rational partition is the key prerequisite for the application of distributed algorithms in distribution networks.This paper proposes community-detection-based approaches to a distribution network partition,includin... A rational partition is the key prerequisite for the application of distributed algorithms in distribution networks.This paper proposes community-detection-based approaches to a distribution network partition,including a non-overlapping partition and a border-node partitioning method.First,a novel electrical distance is defined to quantify the coupling relationships between buses and it is further used as the edge weight in a transformed equivalent graph.Then,a vertex/link partition community detection approach is applied to over-partition the network into high intra-cohesive and low inter-coupled subregions.Following this,a greedy algorithm and a tabu search method are combined to merge these subregions into target numbers according to the scale similarity principle.The proposed approaches take the influence of three-phase imbalance into consideration and they are decoupled from the power flow.Finally,the approaches are tested on an IEEE 123-bus distribution system and the results verify the effectiveness and the credibility of our proposed methods. 展开更多
关键词 Community detection distribution network partition three-phase imbalance
原文传递
RWK-GNN:基于特征增强与子核分解的非平衡图欺诈检测算法
17
作者 于浩淼 刘炜 +2 位作者 孟流畅 陈开睿 宋友 《电子学报》 EI CAS CSCD 北大核心 2024年第10期3382-3391,共10页
金融欺诈对经济和社会稳定造成了严重的威胁,因此开发有效的欺诈检测算法对于保护金融系统的完整性至关重要.目前已有多种基于图学习的欺诈检测算法应用于实际场景当中,这些方法或针对图的结构信息开展分类,或通过图卷积神经网络学习节... 金融欺诈对经济和社会稳定造成了严重的威胁,因此开发有效的欺诈检测算法对于保护金融系统的完整性至关重要.目前已有多种基于图学习的欺诈检测算法应用于实际场景当中,这些方法或针对图的结构信息开展分类,或通过图卷积神经网络学习节点的嵌入式表示进行欺诈检测工作,关注角度相对单一,无法完备地在非平衡多关系图上开展欺诈检测分析.针对以上问题,本论文提出了一种结合随机游走下的特征增强与子核分解的图神经网络欺诈检测算法(Random Walk feature enhancement and Kcore subkernel decomposition Graph Neural Network,RWKGNN),该算法能够高效地挖掘出多关系不平衡图中节点层级与全局网络层级的拓扑信息,并通过子核分解算法优化图结构特征在社区演进角度上的传播与聚合,最终完成欺诈检测与识别.为验证RWK-GNN算法性能,本文使用了图神经网络欺诈检测任务常用的公开数据集进行模型训练与测试.实验结果表明,在同一评价指标下,该方法较相关机器学习算法与图神经网络算法有着较大提升,与CARE-GNN算法相比,该方法的AUC值提升了17%;与PC-GNN算法相比,该方法的AUC值提升了8%;与SIGN算法相比,该方法的AUC值提升了7%. 展开更多
关键词 深度学习 图表示学习 图神经网络 类不平衡 节点分类 金融欺诈检测
下载PDF
基于果蝇算法和卷积神经网络的入侵检测研究
18
作者 王欣源 缪祥华 +2 位作者 黄明巍 张世奇 张世杰 《化工自动化及仪表》 CAS 2024年第5期837-843,863,共8页
提出一种基于果蝇算法(FOA)和卷积神经网络(CNN)的入侵检测方法。FOA实现训练数据的均衡,并结合新型CNN结构(1个输入层、3个卷积层、2个池化层和2个全连接层)改善不均衡数据对入侵检测的影响。研究使用了OpenStack数据集,包含正常数据和... 提出一种基于果蝇算法(FOA)和卷积神经网络(CNN)的入侵检测方法。FOA实现训练数据的均衡,并结合新型CNN结构(1个输入层、3个卷积层、2个池化层和2个全连接层)改善不均衡数据对入侵检测的影响。研究使用了OpenStack数据集,包含正常数据和4类攻击数据。数据特征经处理后转换为灰度图作为CNN的输入数据。实验结果表明:该方法显著提高了入侵检测系统的分类平均精确率,克服了不均衡训练数据的负面影响,实现了CNN在入侵检测中的应用。可为提高入侵检测系统的准确性和可信度提供参考。 展开更多
关键词 入侵检测 FOCNN 数据不平衡 少数类攻击 OPENSTACK CNN新结构
下载PDF
AHLNet:Adaptive Multihead Structure and Lightweight Feature Pyramid Network for Detection of Live Working in Substations
19
作者 Mengle Peng Xiaoyong Jiang +3 位作者 Langyue Huang Zhongyi Li Haiteng Wu Xiaotang Geng 《Machine Intelligence Research》 EI CSCD 2024年第5期983-992,共10页
With the increasing demand for power in society,there is much live equipment in substations,and the safety and standardization of live working of workers are facing challenges.Aiming at these problems of scene complex... With the increasing demand for power in society,there is much live equipment in substations,and the safety and standardization of live working of workers are facing challenges.Aiming at these problems of scene complexity and object diversity in the real-time detection of the live working safety of substation workers,an adaptive multihead structure and lightweight feature pyramid-based network(AHLNet)is proposed in this study,which is based on YOLOV3.First,we take AH-Darknet53 as the backbone network of YOLOV3,which can introduce an adaptive multihead(AMH)structure,reduce the number of network parameters,and improve the feature extraction ability of the backbone network.Second,to reduce the number of convolution layers of the deeper feature map,a lightweight feature pyramid network(LFPN)is proposed,which can perform feature fusion in advance to alleviate the problem of feature imbalance and gradient disappearance.Finally,the proposed AHLNet is evaluated on the datasets of 16 categories of substation safety operation scenarios,and the average prediction accuracy MAP_(50)reaches 82.10%.Compared with YOLOV3,MAP_(50)is increased by 2.43%,and the number of parameters is 90 M,which is only 38%of the number of parameters of YOLOV3.In addition,the detection speed is basically the same as that of YOLOV3,which can meet the real-time and accurate detection requirements for the safe operation of substation staff. 展开更多
关键词 Adaptive multihead structure lightweight feature pyramid substation feature imbalance multiobject detection
原文传递
Coordinated Control of Parallel Three-phase Four-wire Converters in Autonomous AC Microgrids
20
作者 Rui Liu Li Guo Xialin Li 《CSEE Journal of Power and Energy Systems》 SCIE EI CSCD 2024年第5期2065-2078,共14页
The coordinated control of parallel three-phase fourwire converters in autonomous AC microgrids is investigated in this paper.First,based on droop control,virtual impedance is inserted in positive-,negative-and zero-s... The coordinated control of parallel three-phase fourwire converters in autonomous AC microgrids is investigated in this paper.First,based on droop control,virtual impedance is inserted in positive-,negative-and zero-sequences to enhance system damping and imbalance power sharing.Then,to facilitate virtual impedance design,small signal models of the three-sequence equivalent circuits are established respectively.Corresponding indexes are proposed to comprehensively evaluate the impact of sequence virtual impedance on current sharing accuracy,voltage quality at the point of common coupling(PCC)and system stability.In addition,constraint of DClink voltage is also considered to avoid over modulation when subjected to unbalanced loads.Furthermore,to address the PCC voltage degradation resulting from virtual impedance,a voltage imbalance compensation method,based on low-bandwidth communication,is proposed.Finally,simulation and experimental results are provided to verify the correctness of the theory model,indicating that the proposed method can achieve PCC voltage restoration while guaranteeing the current sharing accuracy with desirable dynamics. 展开更多
关键词 AC microgrids sequence virtual impedance three-phase four-wire(3P4W)converter voltage imbalance compensation
原文传递
上一页 1 2 7 下一页 到第
使用帮助 返回顶部