A Tool to Extract Name Entity Recognition From Big Data in Banking Sectors

IF 0.8 4区计算机科学 Q4 COMPUTER SCIENCE, INFORMATION SYSTEMS International Journal of Web Services Research Pub Date : 2020-04-01 DOI:10.4018/IJWSR.2020040102

C. Saju, S. Ravimaran

{"title":"A Tool to Extract Name Entity Recognition From Big Data in Banking Sectors","authors":"C. Saju, S. Ravimaran","doi":"10.4018/IJWSR.2020040102","DOIUrl":null,"url":null,"abstract":"Generally,theInternetistheglobalsystemofinterconnectedcomputernetworks,connectingmillions ofcomputersaswellaspeople,andthusgeneratesamassivequantityofinformationonadailybasis. Thisleadstoextractingthenecessaryinformationusinginformationfiltering(IF)inseveraldomains. Inourimplementation,thenamedentityrecognition(NER)techniqueisemployedtoautomatically extract valuable data from the unstructured natural language texts. As several works has been outlinedindetectingnamedentities,plentyofverydifferentNERtoolsexistforseveraldomains. However,NERremainsagiantchallengesotosolvethisproblemweproposedanovelframeworkby combiningthreeefficientclassifiers.Thisarticleproposesathree-layeredneuralnetworkapproach withconditionalrandomfield(CRF),thePachinkoallocationmodel(PAM),andtheAdaptiveNeuroFuzzyInferenceSystem(ANFIS)fordetectingnamedentitiesinthreesteps.First,aclassifierbased onCRFisemployedtotraintheinputfile.Second,PAMisemployedtoboostthepreviousoutput createdbyCRFtoenhancethelabelannotation.Third,theANFIScapturesthedeepfeaturesofthe informationbyitselffromthepre-trainedinformationtoattainaccuratepredictions.Experimental resultsshowthatthelearnedmodelyieldsabankingdomainwitharecallrateof92%,aprecision rateof95%andF-measureof92%byimplementingitintheRPlatform. KEyWoRD Adaptive Neuro-Fuzzy Inference System, Conditional Random Field, Information Filtering, Named Entity Recognition, Natural Language Processing, Pachinko Allocation Model","PeriodicalId":54936,"journal":{"name":"International Journal of Web Services Research","volume":"47 1","pages":"18-39"},"PeriodicalIF":0.8000,"publicationDate":"2020-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"5","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal of Web Services Research","FirstCategoryId":"94","ListUrlMain":"https://doi.org/10.4018/IJWSR.2020040102","RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"COMPUTER SCIENCE, INFORMATION SYSTEMS","Score":null,"Total":0}

引用次数: 5

Abstract

Generally,theInternetistheglobalsystemofinterconnectedcomputernetworks,connectingmillions ofcomputersaswellaspeople,andthusgeneratesamassivequantityofinformationonadailybasis. Thisleadstoextractingthenecessaryinformationusinginformationfiltering(IF)inseveraldomains. Inourimplementation,thenamedentityrecognition(NER)techniqueisemployedtoautomatically extract valuable data from the unstructured natural language texts. As several works has been outlinedindetectingnamedentities,plentyofverydifferentNERtoolsexistforseveraldomains. However,NERremainsagiantchallengesotosolvethisproblemweproposedanovelframeworkby combiningthreeefficientclassifiers.Thisarticleproposesathree-layeredneuralnetworkapproach withconditionalrandomfield(CRF),thePachinkoallocationmodel(PAM),andtheAdaptiveNeuroFuzzyInferenceSystem(ANFIS)fordetectingnamedentitiesinthreesteps.First,aclassifierbased onCRFisemployedtotraintheinputfile.Second,PAMisemployedtoboostthepreviousoutput createdbyCRFtoenhancethelabelannotation.Third,theANFIScapturesthedeepfeaturesofthe informationbyitselffromthepre-trainedinformationtoattainaccuratepredictions.Experimental resultsshowthatthelearnedmodelyieldsabankingdomainwitharecallrateof92%,aprecision rateof95%andF-measureof92%byimplementingitintheRPlatform. KEyWoRD Adaptive Neuro-Fuzzy Inference System, Conditional Random Field, Information Filtering, Named Entity Recognition, Natural Language Processing, Pachinko Allocation Model

查看原文

微信好友朋友圈 QQ好友复制链接

本刊更多论文

从银行业大数据中提取名称实体识别的工具

一般是theInternetistheglobalsystemofinterconnectedcomputernetworks、connectingmillions ofcomputersaswellaspeople、andthusgeneratesamassivequantityofinformationonadailybasis。Thisleadstoextractingthenecessaryinformationusinginformationfiltering(IF)inseveraldomains。Inourimplementation，thenamedentityrecognition(NER)techniqueisemployedtoautomatically从unstructurednaturallanguage_文本中提取有价值的数据。[as]几个[作品][已经][outlinedindetectingnamedentities，plentyofverydifferentNERtoolsexistforseveraldomains]。然而，NERremainsagiantchallengesotosolvethisproblemweproposedanovelframeworkby combiningthreeefficientclassifiers。Thisarticleproposesathree-layeredneuralnetworkapproach withconditionalrandomfield(CRF)、thePachinkoallocationmodel(PAM)、andtheAdaptiveNeuroFuzzyInferenceSystem(ANFIS)fordetectingnamedentitiesinthreesteps。First，aclassifierbased onCRFisemployedtotraintheinputfile。Second，PAMisemployedtoboostthepreviousoutput createdbyCRFtoenhancethelabelannotation。Third，theANFIScapturesthedeepfeaturesofthe informationbyitselffromthepre-trainedinformationtoattainaccuratepredictions。Experimental resultsshowthatthelearnedmodelyieldsabankingdomainwitharecallrateof92%，aprecision rateof95%andF-measureof92%byimplementingitintheRPlatform。关键词自适应神经模糊推理系统，条件随机场，信息过滤，命名实体识别，自然语言处理，柏青哥分配模型

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文去求助

来源期刊

International Journal of Web Services Research 工程技术-计算机：软件工程

CiteScore

2.40

自引率

0.00%

发文量

审稿时长

>12 weeks

期刊介绍： The International Journal of Web Services Research (IJWSR) is the first refereed, international publication featuring the latest research findings and industry solutions involving all aspects of Web services technology. This journal covers advancements, standards, and practices of Web services, as well as identifies emerging research topics and defines the future of Web services on grid computing, multimedia, and communication. IJWSR provides an open, formal publication for high quality articles developed by theoreticians, educators, developers, researchers, and practitioners for those desiring to stay abreast of challenges in Web services technology.