检索结果-南通市图书馆

Feature extraction for machine learning-based intrusion detection in IoT networks

Digital Communications and Networks 2024年第1期10卷 205-216页

作者： Mohanad Sarhan Siamak Layeghy nour moustafa Marcus Gallagher Marius Portmann University of Queensland BrisbaneQLD4072Australia University of New South Wales CanberraACT2612Australia

A large number of network security breaches in IoT networks have demonstrated the unreliability of current Network Intrusion Detection Systems(NIDSs).Consequently,network interruptions and loss of sensitive data have occurred,which led to an active research area for improving NIDS *** an analysis of related works,it was observed that most researchers aim to obtain better classification results by using a set of untried combinations of Feature Reduction(FR)and Machine Learning(ML)techniques on NIDS ***,these datasets are different in feature sets,attack types,and network ***,this paper aims to discover whether these techniques can be generalised across various *** ML models are utilised:a Deep Feed Forward(DFF),Convolutional Neural Network(CNN),Recurrent Neural Network(RNN),Decision Tree(DT),Logistic Regression(LR),and Naive Bayes(NB).The accuracy of three Feature Extraction(FE)algorithms is detected;Principal Component Analysis(PCA),Auto-encoder(AE),and Linear Discriminant Analysis(LDA),are evaluated using three benchmark datasets:UNSW-NB15,ToN-IoT and *** PCA and AE algorithms have been widely used,the determination of their optimal number of extracted dimensions has been *** results indicate that no clear FE method or ML model can achieve the best scores for all *** optimal number of extracted dimensions has been identified for each dataset,and LDA degrades the performance of the ML models on two *** variance is used to analyse the extracted dimensions of LDA and ***,this paper concludes that the choice of datasets significantly alters the performance of the applied *** believe that a universal(benchmark)feature set is needed to facilitate further advancement and progress of research in this field.

关键词： Feature extraction Machine learning Network intrusion detection system IoT

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Quranic Script Optical Text Recognition Using Deep Learning in IoT Systems

引用

Computers, Materials & Continua 2021年第8期68卷 1847-1858页

作者： Mahmoud Badry Mohammed Hassanin Asghar Chandio nour moustafa School of Engineering and Information Technology UNSW Canberra.ACTNewcastle NSW2620Australia Faculty of Computers and Information Fayoum UniversityFayoumEgypt Quaid-e-Awam University of Engineering Science and TechnologyNawabshahPakistan

Since the worldwide spread of internet-connected devices and rapid advances made in Internet of Things(IoT)systems,much research has been done in using machine learning methods to recognize IoT sensors *** is particularly the case for optical character recognition of handwritten *** text in images has several useful applications,including content-based image retrieval,searching and document *** Arabic language is one of the mostly used tongues in the ***,Arabic text recognition in imagery is still very much in the nascent stage,especially handwritten *** is mainly due to the language complexities,different writing styles,variations in the shape of characters,diacritics,and connected nature of Arabic *** this paper,two deep learning models were *** first model was based on a sequence-to-sequence recognition,while the second model was based on a fully convolution *** measure the performance of these models,a new dataset,called QTID(Quran Text Image Dataset)was *** is the first Arabic dataset that includes Arabic *** consists of 309,720 different 192×64 annotated Arabic word images,which comprise 2,494,428 characters in total taken from the Holy *** annotated images in the dataset were randomly divided into 90%,5%,and 5%sets for training,validation,and testing purposes,*** models were set up to recognize the Arabic Othmani font in the *** results show that the proposed methods achieve state-of-the-art ***,the proposed models surpass expectations in terms of character recognition rate,F1-score,average precision,and recall *** are superior to the best Arabic text recognition engines like Tesseract and ABBYY FineReader.

关键词： OCR quranic script IoT deep learning

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Privacy-preserved learning from non-i.i.d data in fog-assisted IoT:A federated learning approach

引用

Digital Communications and Networks 2024年第2期10卷 404-415页

作者： Mohamed Abdel-Basset Hossam Hawash nour moustafa Imran Razzak Mohamed Abd Elfattah Faculty of Computers and Informatics Zagazig UniversityZagazigSharqiyah44519Egypt School of Engineering and Information Technology University of New South Wales@ADFACanberraACT2600Australia Deakin University Geelong Waurn Ponds CampusAustralia Computer Science Department Misr Higher Institute for Commerce and ComputersMansoura35511Egypt

With the prevalence of the Internet of Things(IoT)systems,smart cities comprise complex networks,including sensors,actuators,appliances,and cyber *** complexity and heterogeneity of smart cities have become vulnerable to sophisticated cyber-attacks,especially privacy-related attacks such as inference and data poisoning *** Learning(FL)has been regarded as a hopeful method to enable distributed learning with privacypreserved intelligence in IoT *** though the significance of developing privacy-preserving FL has drawn as a great research interest,the current research only concentrates on FL with independent identically distributed(i.i.d)data and few studies have addressed the non-i.i.d *** is known to be vulnerable to Generative Adversarial Network(GAN)attacks,where an adversary can presume to act as a contributor participating in the training process to acquire the private data of other *** paper proposes an innovative Privacy Protection-based Federated Deep Learning(PP-FDL)framework,which accomplishes data protection against privacy-related GAN attacks,along with high classification rates from non-i.i.d ***-FDL is designed to enable fog nodes to cooperate to train the FDL model in a way that ensures contributors have no access to the data of each other,where class probabilities are protected utilizing a private identifier generated for each *** PP-FDL framework is evaluated for image classification using simple convolutional networks which are trained using MNIST and CIFAR-10 *** empirical results have revealed that PF-DFL can achieve data protection and the framework outperforms the other three state-of-the-art models with 3%–8%as accuracy improvements.

关键词： Privacy preservation Federated learning Deep learning Fog computing Smart cities

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Enhancing IoT anomaly detection performance for federated learning

引用

Digital Communications and Networks 2022年第3期8卷 314-323页

作者： Brett Weinger Jinoh Kim Alex Sim Makiya Nakashima nour moustafa K.John Wu Stony Brook University Stony BrookNYUSA Lawrence Berkeley National Laboratory BerkeleyCAUSA Texas A&M University CommerceTXUSA University of New South Wales CanberraAustralia

Federated Learning (FL) with mobile computing and the Internet of Things (IoT) is an effective cooperative learning approach. However, several technical challenges still need to be addressed. For instance, dividing the training process among several devices may impact the performance of Machine Learning (ML) algorithms, often significantly degrading prediction accuracy compared to centralized learning. One of the primary reasons for such performance degradation is that each device can access only a small fraction of data (that it generates), which limits the efficacy of the local ML model constructed on that device. The performance degradation could be exacerbated when the participating devices produce different classes of events, which is known as the class balance problem. Moreover, if the participating devices are of different types, each device may never observe the same types of events, which leads to the device heterogeneity problem. In this study, we investigate how data augmentation can be applied to address these challenges and improving detection performance in an anomaly detection task using IoT datasets. Our extensive experimental results with three publicly accessible IoT datasets show the performance improvement of up to 22.9% with the approach of data augmentation, compared to the baseline (without relying on data augmentation). In particular, stratified random sampling and uniform random sampling show the best improvement in detection performance with only a modest increase in computation time, whereas the data augmentation scheme using Generative Adversarial Networks is the most time-consuming with limited performance benefits.

关键词： Data augmentation Federated learning Internet of things Anomaly detection Machine learning

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Few-Shot Learning for Discovering Anomalous Behaviors in Edge Networks

引用

Computers, Materials & Continua 2021年第11期69卷 1823-1837页

作者： Merna Gamal Hala M.Abbas nour moustafa Elena Sitnikova Rowayda A.Sadek Department of Information Technology Helwan UniversityCairoEgypt Department of Computer Science Helwan UniversityCairoEgypt School of Engineering and Information Technology University of New South Wales at ADFACanberraAustralia

Intrusion Detection Systems(IDSs)have a great interest these days to discover complex attack events and protect the critical infrastructures of the Internet of Things(IoT)*** IDSs based on shallow and deep network architectures demand high computational resources and high volumes of data to establish an adaptive detection engine that discovers new families of attacks from the edge of IoT ***,attackers exploit network gateways at the edge using new attacking scenarios(i.e.,zero-day attacks),such as ransomware and Distributed Denial of Service(DDoS)*** paper proposes new IDS based on Few-Shot Deep Learning,named CNN-IDS,which can automatically identify zero-day attacks from the edge of a network and protect its IoT *** proposed system comprises two-methodological stages:1)a filtered Information Gain method is to select the most useful features from network data,and 2)one-dimensional Convolutional Neural Network(CNN)algorithm is to recognize new attack types from a network’s *** proposed model is trained and validated using two datasets of the UNSW-NB15 and *** experimental results showed that it enhances about a 3%detection rate and around a 3%–4%falsepositive rate with the UNSW-NB15 dataset and about an 8%detection rate using the BoT-IoT dataset.

关键词： Convolution neural network information gain few-shot learning IoT edge computing

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

A Secured Message Transmission Protocol for Vehicular Ad Hoc Networks

引用

Computers, Materials & Continua 2021年第7期68卷 229-246页

作者： A.F.M.Suaib Akhter A.F.M.Shahen Shah Mohiuddin Ahmed nour moustafa UnalÇavu¸soglu Ahmet Zengin Department of Computer Engineering Sakarya UniversitySakarya54050Turkey Department of Electrical and Electronics Engineering Istanbul Gelisim UniversityIstanbulTurkey School of Science Edith Cowan UniversityPerthWA6027Australia School of Engineering and Information Technology UNSWCanberra BC2610Australia

Vehicular Ad hoc Networks(VANETs)become a very crucial addition in the Intelligent Transportation System(ITS).It is challenging for a VANET system to provide security services and parallelly maintain high throughput by utilizing limited *** overcome these challenges,we propose a blockchain-based Secured Cluster-based MAC(SCB-MAC)*** nearby vehicles heading towards the same direction will form a cluster and each of the clusters has its blockchain to store and distribute the safety *** message which contains emergency information and requires Strict Delay Requirement(SDR)for transmission are called safety messages(SM).Cluster Members(CMs)sign SMs with their private keys while sending them to the blockchain to confirm authentication,integrity,and confidentiality of the message.A Certificate Authority(CA)is responsible for physical verification,key generation,and privacy preservation of the *** implemented a test scenario as proof of concept and tested the safety message transmission(SMT)protocol in a real-world *** and storage overhead analysis shows that the proposed protocol for SMT implements security,authentication,integrity,robustness,non-repudiation,*** maintaining the *** that are less important compared to the SMs are called non-safety messages(NSM)and vehicles use RTS/CTS mechanism for NSM *** studies show that the proposed NSM transmission method maintains 6 times more throughput,2 times less delay and 125%less Packet Dropping Rate(PDR)than traditional MAC *** results prove that the proposed protocol outperforms the traditional MAC protocols.

关键词： Ad hoc networks data security digital signatures distributed storage intelligent vehicles vehicular ad hoc networks wireless communication

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Association of anxiety- depressive disorders with irritable bowel syndrome among patients attending a rural family practice center: a comparative cross- sectional study

引用

General Psychiatry 2021年第6期34卷 16-23页

作者： Alaa Adel Mohammed Heba Ahmed moustafa Hebatallah nour-Eldein Rabab Atta Saudi Family Medicine Department Suez Canal UniversityIsmailiaEgypt

Background Comorbidity of irritable bowel syndrome(IBS)and psychiatric disorders is common,and the prevalence of at least one psychiatric disorder has been reported as high as 80%among patients with *** To explore the association of anxiety-depressive disorders with IBS and its different subtypes,and to evaluate the associations of lifestyle habits,dietary habits and sleeping quality with *** A comparative cross-sectional study was conducted at the AL-Mahsama Family Practice Center,Ismailia,*** was carried out between October 2019 and October *** were categorised into 175 patients with IBS,diagnosed using the Rome IV criteria,and 175 patients without IBS.A semistructured questionnaire was used to collect data on sociodemographic characteristics,lifestyle habits,dietary habits and sleep quality from both *** Hamilton Anxiety Rating Scale was used to assess anxiety symptoms,whereas the Beck Depression Inventory Second Edition was used to assess depression *** There was a high statistically significant difference between both groups with regard to age,education,occupation and socioeconomic status(SES),being a smoker,being physically inactive,having sleep disturbance and having irregular meals;being either obese or overweight was more reported in the IBS *** was a high statistically significant difference in the rate of anxiety and depression between patients with and without ***,moderate and severe anxiety were reported in 37.1%,42.9%and 20.0%of patients with IBS while most(80.0%)of the patients without IBS reported mild *** depression,mild,moderate and severe depression were reported in 60.0%,14.3%and 25.7%of the patients with IBS while most(82.9%)of the non-IBS participants reported mild *** The study shows a significant association between anxiety-depressive disorders and IBS,but no significant associations between anxiety-depressive disorders and IBS subtypes.

关键词： patients disorders sectional

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Involvement of Quebracho tannins in diet alters productive and reproductive efficiency of postpartum buffalo cows

引用

Animal Nutrition 2019年第1期5卷 80-86页

作者： Sobhy M.A.Sallam Marwa F.A.Attia Adel N.M.nour El-Din Samir Z.El-Zarkouny Adel M.Saber Hani M.El-Zaiat moustafa M.Zeitoun Department of Animal and Fish Production Faculty of AgricultureAlexandria University Animal Production Research Institute Agricultural Research CenterDokki

This study was conducted to investigate the effects of 10 weeks supplementation of Quebracho tannins(QT; 0 [control], 100 [QT100] or 200 g/[cow$d] [QT_(200)]) to 30 multiparous postpartum buffalo cows(10 cows per group) on milk yield and composition, blood metabolites and reproductive *** of QT100 had no significant effect on milk yield, whereas QT_(200) decreased(P < 0.05) this trait. Compared with the control group, both QT levels decreased(P < 0.05) fat-corrected milk(FCM)yield, but no significant effects were found on percentages of milk fat and protein. Contrariwise, yields of milk fat, lactose and milk protein were decreased(P < 0.05) when QT_(200) was supplemented. The solids nonfat(SNF) percentage and yield were decreased(P < 0.05) with QT100 supplementation. Moreover, QT tended to numerically reduce total number of ovarian follicles, number of small follicles, peripheral progesterone concentration and conception rate. Supplementation of QT_(200) numerically increased number of large follicles, mean diameter of large follicle, number and diameters of corpora lutea. The inclusion of QT_(200) shortened days open(DO) and decreased number of services per ***, QT did not show significant effects on serum total protein, albumin, globulin, glucose,cholesterol and triglycerides concentrations. Supplementation of QT100 caused an increase(P < 0.05) of serum urea compared with that in control and QT_(200) groups. Generally, QT decreased(P < 0.05) serum creatinine concentration. Therefore, the supplementation of a commercial QT to early lactating Egyptian buffalo cows displayed negative consequences on their productive and reproductive performances.

关键词： Buffalo Tannins Milk traits Blood metabolites Ovarian follicles Conception rate

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Impact of Quebracho Tannins Supplementation on Productive and Reproductive Efficiency of Dairy Cows

引用

Open Journal of Animal Sciences 2016年第4期6卷 269-288页

作者： Marwa F. A. Attia Adel N. M. nour El-Din Samir Z. El-Zarkouny Hani M. El-Zaiat moustafa M. Zeitoun Sobhy M. A. Sallam Department of Animal and Fish Production Faculty of Agriculture Alexandria University Alexandria Egypt Department of Arid Land Agriculture Faculty of Meteorology Environment and Arid Land Agriculture Jeddah Saudi Arabia Department of Animal Production and Breeding College of Agriculture and Veterinary Medicine Qassim University Qassim Saudi Arabia

This study was conducted to investigate effects of supplementing two doses of quebracho tannins (QT;100 or 200 g/cow/d;QT100 or QT200) pre and post parturition to thirty Holstein dairy cows on Dry Matter Intake (DMI), milk yield and composition, blood metabolites and reproductive performance for 12 weeks. There were no significant changes in DMI due to QT at transition period of dairy cows. QT supplementation at either level decreased (P 0.05) the total number of ovarian follicles, number of large follicles, diameter of largest follicle, number of small follicles, number and diameter of corpus luteum, and progesterone concentration. The inclusion of QT increased days open and number of services per conception, which consequently decreased conception rate, compared to the control cows. QT100 decreased (P

关键词： Quebracho Tannin Feed Intake Milk Yield Follicular Dynamics Conception Rate

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

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：