英文作文跑题智能化检测技术及应用  

Intelligent Detection Technology and Application of English Composition Digression

作　　者：刘娜 LIU Na(College of Foreign Languages,Xianyang Normal University,Xianyang 712000,China)

机构地区：[1]咸阳师范学院外国语学院,陕西咸阳712000

出　　处：《微型电脑应用》2021年第9期182-184,共3页Microcomputer Applications

摘　　要：现有的英文作文辅助评分系统大多缺乏有效的跑题检测手段,为此提出并设计了一种英文作文跑题智能化检测技术,利用LDA模型获取作文的主题信息并通过Word2vec训练模型得到准确表达词义的词向量,基于二者的综合作用效果定位跑题的文档。对比实验结果表明,所提出的算法,综合评价指标F值为91.2%,具有非常好的跑题检测能力,对于英语教学能够发挥优异的辅助功能。Most of the existing English composition assistant scoring systems lack of effective means to detect off topic.Therefore,this paper proposes and designs an intelligent detection algorithm for English composition digression.LDA model is used to obtain the topic information of the composition,and the Word2vec training model is used to obtain the word vector to accurately express the word meaning.Based on the comprehensive effect of the two,the off topic document is located.The experimental results show that the comprehensive evaluation index F value of the algorithm proposed in this paper is 91.2%,which has a very outstanding ability of off topic detection,and can play an excellent auxiliary function for English teaching.

关 键 词：智能化检测 作文评分系统 LDA主题模型 Word2vec模型 

分 类 号：H315[语言文字—英语]