Computer and Modernization ›› 2018, Vol. 0 ›› Issue (07): 53-.doi: 10.3969/j.issn.1006-2475.2018.07.011

Previous Articles     Next Articles

LimitedDomainKnowledgeQuestionAnsweringSystemBasedonBI-LSTM-CRF

  

  1. (SchoolofComputerScienceandTechnology,TaiyuanUniversityofScienceandTechnology,Taiyuan030024,China)
  • Received:2018-01-08 Online:2018-08-23 Published:2018-08-27

Abstract: Withthedevelopmentofopenfieldquestion-answeringsystemandtheurgentneedoftheintegrationofmechanicalindustryandartificialintelligence,itisnecessarytoestablishaknowledgebasequestionansweringsystemformachineryfield.Basedonmechanicalindustrydataandnaturallanguageprocessingtechniques,anetworkmodelbasedonconditionalrandomfieldandlongandshorttermmemoryneuralnetworkisproposedtoimprovetheinformationextractionperformanceandtoestablishaknowledgebasequestion-answeringsysteminmachineryindustry.Throughthecomparativeanalysisofexperimentaldata,themodelhasachievedgoodresults.

Key words: CRF, LSTM, informationextraction, questionanswering

CLC Number: