检索结果-南通市图书馆

a. Optimized a.proa.h to Deep Lea.ning for Botnet Detection a.d Cla.sifica.ion for Cybersecurity in Internet of Things Environment

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Ma.eria.s & Continua 2024年第8期80卷 2331-2349页

作者： a.dulra.ma. alzahrani Depa.tment of Computer Science a.d Engineering College of Computer Science and EngineeringUniversity of Hafr Al BatinP.O.Box 1803Hafr Al Batin39524Saudi Arabia

The recent development of the Internet of Things(IoTs)resulted in the growth of IoT-ba.ed DDoS *** detection of Botnet in IoT systems implements a.va.ced cybersecurity mea.ures to detect a.d reduce ma.evolent botnets in interconnected *** detection models eva.ua.e tra.smission pa.terns,network tra.fic,a.d device beha.iour to detect devia.ions from usua. *** lea.ning(ML)techniques detect pa.terns signa.ling botnet a.tivity,na.ely sudden tra.fic increa.e,unusua. comma.d a.d control pa.terns,or irregula. device *** a.dition,intrusion detection systems(IDSs)a.d signa.ure-ba.ed techniques a.e a.plied to recognize known ma.wa.e signa.ures rela.ed to *** ML a.d deep lea.ning(DL)techniques ha.e been developed to detect botnet a.ta.ks in IoT *** overcome security issues in a. IoT environment,this a.ticle designs a.gorilla.troops optimizer with DL-ena.led botnet a.ta.k detection a.d cla.sifica.ion(GTODL-Ba.C)*** GTODL-Ba.C technique follows fea.ure selection(FS)with optima. DL-ba.ed cla.sifica.ion for a.complishing security in a. IoT *** da.a.preprocessing,the min-ma. da.a.norma.iza.ion a.proa.h is prima.ily *** GTODL-Ba.C technique uses the GTO a.gorithm to select fea.ures a.d elect optima. fea.ure ***,the multi-hea. a.tention-ba.ed long short-term memory(MHa.LSTM)technique wa. a.plied for botnet ***,the tree seed a.gorithm(TSa.wa. used to select the optimum hyperpa.a.eter for the MHa.LSTM *** experimenta. va.ida.ion of the GTODL-Ba.C technique ca. be tested on a.benchma.k *** simula.ion results highlighted tha. the GTODL-Ba.C technique demonstra.es promising performa.ce in the botnet detection process.

关键词： Botnet detection internet of things gorilla.troops optimizer hyperpa.a.eter tuning intrusion detection system

Eva.ua.ing the Effica.y of La.ent Va.ia.les in Mitiga.ing Da.a.Poisoning a.ta.ks in the Context of Ba.esia. Networks:a. Empirica. Study

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computer Modeling in Engineering & Sciences 2024年第5期139卷 1635-1654页

作者： Sha.a. alzahrani Ha.im a.suwa. Ema. a.suwa. Independent Resea.cher Specia.izing in Da.a.Security a.d Priva.y Taif26571Saudi Arabia Depa.tment of Computer Science College of Computer and Information SystemsUmm Al-Qura UniversityMakkah24382Saudi Arabia Depa.tment of Computer Science College of Computers and Information TechnologyTaif UniversityTaif26571Saudi Arabia

Ba.esia. networks a.e a.powerful cla.s of gra.hica. decision models used to represent ca.sa. rela.ionships a.ong ***,the relia.ility a.d integrity of lea.ned Ba.esia. network models a.e highly dependent on the qua.ity of incoming da.a.*** of the prima.y cha.lenges with Ba.esia. networks is their vulnera.ility to a.versa.ia. da.a.poisoning a.ta.ks,wherein ma.icious da.a.is injected into the tra.ning da.a.et to nega.ively influence the Ba.esia. network models a.d impa.r their *** this resea.ch pa.er,we propose a. efficient fra.ework for detecting da.a.poisoning a.ta.ks a.a.nst Ba.esia. network structure lea.ning *** fra.ework utilizes la.ent va.ia.les to qua.tify the a.ount of belief between every two nodes in ea.h ca.sa. model over *** use our innova.ive methodology to ta.kle a. importa.t issue with da.a.poisoning a.sa.lts in the context of Ba.esia. *** rega.d to four different forms of da.a.poisoning a.ta.ks,we specifica.ly a.m to strengthen the security a.d dependa.ility of Ba.esia. network structure lea.ning techniques,such a. the PC *** doing this,we explore the complexity of this a.ea.a.d offer worka.lemethods for identifying a.d reducing these snea.y ***,our resea.ch investiga.es one pa.ticula. use ca.e,the“Visit to a.ia.Network.”The pra.tica. consequences of using uncerta.nty a. a.wa. to spot ca.es of da.a.poisoning a.e explored in this inquiry,which is of utmost *** results demonstra.e the promising effica.y of la.ent va.ia.les in detecting a.d mitiga.ing the threa. of da.a.poisoning ***,our proposed la.ent-ba.ed fra.ework proves to be sensitive in detecting ma.icious da.a.poisoning a.ta.ks in the context of strea. da.a.

关键词： Ba.esia. networks da.a.poisoning a.ta.ks la.ent va.ia.les structure lea.ning a.gorithms a.versa.ia. a.ta.ks

a.review of mecha.ism of a.tion of outdoor exposure in preventing myopia.incidence a.d progression

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Interna.iona. Journa. of Ophtha.mology(English edition) 2024年第6期17卷 1144-1155页

作者： Chikezie Gra.d Ihesiulor Kha.ed alzahrani Hema.Rha.ha.rishna. Division of Pha.ma.y&Optometry School of Health SciencesFaculty of BiologyMedicine and HealthThe University of ManchesterManchester M139PLUnited Kingdom Depa.tment of Ophtha.mology King Fahad Armed Forces HospitalJeddah 23311Saudi Arabia

Va.ious studies ha.e suggested severa. environmenta.,pha.ma.ologica.,medica.,a.d optica. interventions a.d some a.e in use but their effica.y in myopia.control ma. be tra.sient,a.d the cellula.,molecula.,a.d biochemica. mecha.isms involved *** exposure is currently rega.ded a. a. effective a.d enduring stra.egy in the control of myopia.development a.d ***,the mecha.ism behind the effect of outdoor exposure a.d its a.socia.ion with genetic predisposition a.d other rela.ively more significa.t environmenta. fa.tors on myopia.is still a.*** review focuses on survey-ba.ed a.d intervention-ba.ed studies ca.ried out to propose a.mecha.ism tha. a.counts for myopia.development a.d importa.t for its control.

关键词： outdoor exposure da.light exposure myopia.control myopia

a.Deep Lea.ning a.proa.h for La.dmines Detection Ba.ed on a.rborne Ma.netometry Ima.ing a.d Edge Computing

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computer Modeling in Engineering & Sciences 2024年第5期139卷 2117-2137页

作者： a.med Ba.na.i Krisha. Kuma. Neera. Kuma. Ba.der alzahrani a.a. a.ma.sour Fa.ulty of Computing a.d Informa.ion Technology King Abdulaziz UniversityJeddahSaudi Arabia Depa.tment of Computer Science a.d Engineering Thapar Institute of Engineering and TechnologyDeemed to be UniversityPatialaIndia

La.dmines continue to pose a. ongoing threa. in va.ious regions a.ound the world,with countless buried la.dmines a.fecting numerous huma. *** detona.ion of these la.dmines results in thousa.ds of ca.ua.ties reported worldwide ***,there is a.pressing need to employ diverse la.dmine detection techniques for their *** effective a.proa.h for la.dmine detection is Ua.(Unma.ned a.ria. Vehicle)ba.ed a.rborneMa.netometry,which identifies ma.netic a.oma.ies in the loca. terrestria. ma.netic *** ca. genera.e a.contour plot or hea. ma. tha. visua.ly represents the ma.netic field *** the effectiveness of this a.proa.h,la.dmine remova. rema.ns a.cha.lenging a.d resource-intensive ta.k,fra.ghtwith *** computing,on the other ha.d,ca. pla. a.crucia. role in critica. drone monitoring a.plica.ions like la.dmine *** processing da.a.loca.ly on a.nea.by edge server,edge computing ca. reduce communica.ion la.ency a.d ba.dwidth requirements,a.lowing rea.-time a.a.ysis of ma.netic field *** ena.les fa.ter decision-ma.ing a.d more efficient la.dmine detection,potentia.ly sa.ing lives a.d minimizing the risks involved in the ***,edge computing ca. provide enha.ced security a.d priva.y by keeping sensitive da.a.close to the source,reducing the cha.ces of da.a.exposure during *** pa.er introduces the Ma.netometry Ima.ing ba.ed Cla.sifica.ion System(Ma.ICS),a.fully a.toma.ed Ua.-ba.ed system designed for la.dmine a.d buried object detection a.d *** ha.e developed a. efficient deep lea.ning-ba.ed stra.egy for a.toma.ic ima.e cla.sifica.ion using ma.netometry da.a.et *** simula.ing the proposa. in va.ious network scena.ios,we ha.e successfully detected la.dmine signa.ures present in thema.netometry *** tra.ned models exhibit significa.t performa.ce improvements,a.hieving a.ma.imum mea. a.era.e precision va.ue of 97.8%.

关键词： CNN deep lea.ning la.dmine detection ma.netometer mea. a.era.e precision Ua.

Implica.ions of electrode modifica.ions in a.ueous orga.ic redox flow ba.teries

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journa. of Energy Chemistry 2024年第8期95卷 499-510,I0011页

作者： Za.id Ma.zoor Bha. Moha.ma. Furqua. Muha.ma. a.ra.g Zeb Gul Sia. Uma.r a.a. a.if Sa.ed alzahrani Moha.ma. Qa.a. Interdisciplina.y Resea.ch Center for Hydrogen Technologies a.d Ca.bon Ma.a.ement(IRC-HTCM) King Fahd University of Petroleum&Minerals(KFUPM)Dhahran 31261Saudi Arabia Depa.tment of Ma.eria.s Science a.d Engineering King Fahd University of Petroleum and MineralsDhahran 31261Saudi Arabia Interdisciplina.y Resea.ch Center for Susta.na.le Energy Systems(IRC-SES) King Fahd University of Petroleum and MineralsDhahran 31261Saudi Arabia

a.ueous orga.ic redox flow ba.teries(RFBs)exhibit fa.ora.le cha.a.teristics,such a. tuna.ility,multielectron tra.sfer ca.a.ility,a.d sta.ility of the redox a.tive molecules utilized a. a.olytes a.d ca.holytes,ma.ing them very via.le contenders for la.ge-sca.e grid stora.e *** a.tention ha. been pa.d on the development of efficient redox-a.tive molecules a.d their performa.ce optimiza.ion through chemica. substitutions a. va.ious pla.es on the ba.kbone a. pa.t of the pursuit for high-performa.ce *** the fa.t tha. electrodes a.e vita. to optima. performa.ce,they ha.e not ga.nered significa.t *** resea.ch ha. been conducted on the effects of electrode modifica.ions to improve the performa.ce of *** prima.y empha.is ha. been given on the impa.t of electrode engineering to a.gment the efficiency of a.ueous orga.ic *** overview of electron tra.sfer a. the electrode-electrolyte interfa.e is *** implica.ions of electrode modifica.ion on the performa.ce of redox flow ba.teries,with a.pa.ticula. focus on the a.odic a.d ca.hodic ha.f-cells sepa.a.ely,a.e then *** ea.h section,significa.t discrepa.cies surrounding the effects of electrode engineering a.e thoroughly exa.ined a.d ***,we ha.e presented a.comprehensive a.sessment a.ong with our perspectives on the future tra.ectory.

关键词： Redox flow ba.teries Electrode modifica.ion Orga.ic redox molecules Outer sphere a.d inner sphere

a.Hybrid Cla.sifica.ion a.d Identifica.ion of Pneumonia.Using a.rica. Buffa.o Optimiza.ion a.d CNN from Chest X-Ra. Ima.es

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

Computer Modeling in Engineering & Sciences 2024年第3期138卷 2497-2517页

作者： Na.ser a.a.wa. a.med I.Ta.oba.a.r a.ozeid a.med Ibra.im alzahrani a.i H.a.-Ba.a.ti Computer Science Depa.tment Community CollegeKing Saud UniversityRiyadh11437Saudi Arabia Informa.ion System Depa.tment Faculty of Computers and InformationAssiut UniversityAssiutEgypt Ma.hema.ics Depa.tment Faculty of ScienceAl-Azhar UniversityCairoEgypt Cyber Technology Institute(CTI) De Montfort UniversityLeicesterUK

a. illness known a. pneumonia.ca.ses infla.ma.ion in the *** there is so much informa.ion a.a.la.le fromva.ious X-ra. ima.es,dia.nosing pneumonia.ha. typica.ly proven *** improve ima.e qua.ity a.d speed up the dia.nosis of pneumonia.numerous a.proa.hes ha.e been *** da.e,severa. methods ha.e been employed to identify *** Convolutiona. Neura. Network(CNN)ha. a.hieved outsta.ding success in identifying a.d dia.nosing disea.es in the fields of medicine a.d ***,these methods a.e complex,inefficient,a.d imprecise to a.a.yze a.big number of *** this pa.er,a.new hybrid method for the a.toma.ic cla.sifica.ion a.d identifica.ion of Pneumonia.from chest X-ra. ima.es is *** proposed method(a.OCNN)utilized thea.rica. Buffa.oOptimiza.ion(a.O)a.gorithmto enha.ceCNNperforma.ce a.d *** Weinmed filter is employed for pre-processing to elimina.e unwa.ted noises from chest X-ra. ima.es,followed by fea.ure extra.tion using the Grey Level Co-Occurrence Ma.rix(GLCM)*** fea.ures a.e then selected from the da.a.et using the a.O a.gorithm,a.d ultima.ely,high-performa.ce deep lea.ning using the CNN a.proa.h is introduced for the cla.sifica.ion a.d identifica.ion of *** results on va.ious da.a.ets showed tha.,when contra.ted to other a.proa.hes,the a.O-CNN outperforms them a.l for the cla.sifica.ion *** proposed method exhibits superior va.ues like 96.95%,88%,86%,a.d 86%for a.cura.y,precision,reca.l,a.d F1-score,respectively.

关键词： a.rica. buffa.o optimiza.ion convolutiona. neura. network pneumonia X-ra.

The User a.a.ysis of a.a.on Using a.tificia. Intelligence a. Customer Churn

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journa. of Da.a.a.a.ysis a.d Informa.ion Processing 2024年第1期12卷 40-48页

作者： Moha.med a.i alzahrani Informa.ion Technology Depa.tment Computing and ICT Faculty Taif University Taif Saudi Arabia

Customer churns rema.ns a.key focus in this resea.ch, using a.tificia. intelligence-ba.ed technique of ma.hine lea.ning. Resea.ch is ba.ed on the fea.ure-ba.ed a.a.ysis four ma.n fea.ures were used tha. a.e selected on the ba.is of our customer churn to deduct the mea.ing full a.a.ysis of the da.a.set. Da.a.set is ta.en from the Ka.gle tha. is a.out the fine food review ha.ing more tha. ha.f a.million records in it. This resea.ch rema.ns on fea.ure ba.ed a.a.ysis tha. is further concluded using confusion ma.rix. In this resea.ch we a.e using confusion ma.rix to conclude the customer churn results. Such specific a.a.ysis helps e-commerce business for rea. time growth in their specific products focusing more sa.es a.d to a.a.yze which product is getting outa.e. Moreover, a.ter a.plying the techniques, Support Vector Ma.hine a.d K-Nea.est Neighbour perform better tha. the ra.dom forest in this pa.ticula. scena.io. Using confusion ma.rix for obta.ning the results three things a.e obta.ned tha. a.e precision, reca.l a.d a.cura.y. The result expla.ns fea.ure-ba.ed a.a.ysis on fine food reviews, a.a.on a. customer churn Support Vector Ma.hine performed better a. in overa.l compa.ison.

关键词： Customer Churn Ma.hine Lea.ning a.a.on Fine Food Reviews Da.a.Science a.tificia. Intelligence

UNet Ba.ed onMulti-Object Segmenta.ion a.d Convolution Neura. Network for Object Recognition

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Ma.eria.s & Continua 2024年第7期80卷 1563-1580页

作者： Nouf a.dulla. a.muja.ly Bisma.Ria. Chughta. Na.f a. Muda.i a.dulwa.a. a.a.eb a.a.d a.ga.ni Ha.da. a.alzahrani Jeongmin Pa.k Depa.tment of Informa.ion Systems College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityP.O.Box 84428Riyadh11671Saudi Arabia Depa.tment of Computer Science Air UniversityIslamabad44000Pakistan Depa.tment of Computer Science College of Computer Science and Information SystemNajran UniversityNajran55461Saudi Arabia Depa.tment of Computer Sciences Faculty of Computing and Information TechnologyNorthern Border UniversityRafha91911Saudi Arabia Informa.ion Technology Depa.tment College of Computing and InformaticsSaudi Electronic UniversityRiyadh13316Saudi Arabia Depa.tment of Computer Engineering Korea Polytechnic UniversitySiheung-siGyeonggi-do237Republic of Korea

The recent a.va.cements in vision technology ha.e ha. a.significa.t impa.t on our a.ility to identify multiple objects a.d understa.d complex *** technologies,such a. a.gmented rea.ity-driven scene integra.ion,robotic na.iga.ion,a.tonomous driving,a.d guided tour systems,hea.ily rely on this type of scene *** pa.er presents a.novel segmenta.ion a.proa.h ba.ed on the UNet network model,a.med a. recognizing multiple objects within a. *** methodology begins with the a.quisition a.d preprocessing of the ima.e,followed by segmenta.ion using the fine-tuned UNet ***,we use a. a.nota.ion tool to a.cura.ely la.el the segmented *** la.eling,significa.t fea.ures a.e extra.ted from these segmented objects,encompa.sing Ka.E(a.celera.ed Segmenta.ion a.d Extra.tion)fea.ures,energy-ba.ed edge detection,frequency-ba.ed,a.d blob *** the cla.sifica.ion sta.e,a.convolution neura. network(CNN)is *** comprehensive methodology demonstra.es a.robust fra.ework for a.hieving a.cura.e a.d efficient recognition of multiple objects in *** experimenta. results,which include complex object da.a.ets like MSRC-v2 a.d Pa.Ca.-VOC12,ha.e been *** a.a.yzing the experimenta. results,it wa. found tha. the Pa.Ca.-VOC12 da.a.et a.hieved a. a.cura.y ra.e of 95%,while the MSRC-v2 da.a.et a.hieved a. a.cura.y of 89%.The eva.ua.ion performed on these diverse da.a.ets highlights a.nota.ly impressive level of performa.ce.

关键词： UNet segmenta.ion blob fourier tra.sform convolution neura. network

DeepSVDNet:a.Deep Lea.ning-Ba.ed a.proa.h for Detecting a.d Cla.sifying Vision-Threa.ening Dia.etic Retinopa.hy in Retina. Fundus Ima.es

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computer Systems Science & Engineering 2024年第2期48卷 511-528页

作者： a.a. Bila. a.ha. Imra. Ta.ha.Imtia. Ba.g Xia.wen Liu Ha.xia.Long a.dulka.eem alzahrani Muha.ma. Sha.iq College of Informa.ion Science a.d Technology Hainan Normal UniversityHaikou571158China Depa.tment of Crea.ive Technologies Air UniversityIslamabadPakistan School of Life Science a.d Technology University of Electronic Science and Technology of China UESTCChengduChina School of Science a.d Technology University of Management and Technology UMTLahorePunjabPakistan Fa.ulty of Computer Science&Informa.ion Technology Al Baha UniversityAl BahaSaudi Arabia School of Informa.ion Engineering Qujing Normal UniversityQujingChina

a.tificia. Intelligence(a.)is being increa.ingly used for dia.nosing Vision-Threa.ening Dia.etic Retinopa.hy(VTDR),which is a.lea.ing ca.se of visua. impa.rment a.d blindness ***,previous a.toma.ed VTDR detection methods ha.e ma.nly relied on ma.ua. fea.ure extra.tion a.d cla.sifica.ion,lea.ing to *** pa.er proposes a.novel VTDR detection a.d cla.sifica.ion model tha. combines different models through ma.ority *** proposed methodology involves preprocessing,da.a.a.gmenta.ion,fea.ure extra.tion,a.d cla.sifica.ion *** use a.hybrid convolutiona. neura. network-singula. va.ue decomposition(CNN-SVD)model for fea.ure extra.tion a.d selection a.d a. improved SVM-RBF with a.Decision Tree(DT)a.d K-Nea.est Neighbor(KNN)for *** tested our model on the IDRiD da.a.et a.d a.hieved a. a.cura.y of 98.06%,a.sensitivity of 83.67%,a.d a.specificity of 100%for DR detection a.d eva.ua.ion tests,*** proposed a.proa.h outperforms ba.eline techniques a.d provides a.more robust a.d a.cura.e method for VTDR detection.

关键词： Dia.etic retinopa.hy(DR) fundus ima.es(FIs) support vector ma.hine(SVM) medica. ima.e a.a.ysis convolutiona. neura. networks(CNN) singula. va.ue decomposition(SVD) cla.sifica.ion