基于文档分布式表示的英语作文跑题分析模型  

Off-topic English essay detection model based on distributed representation of documents

作　　者：刘剑 黄桂敏[1] 潘婷婷 LIU Jian;HUANG Guimin;PAN Tingting(School of Information and Communication Engineering,Guilin University of Electronic Technology,Guilin 541004,China)

机构地区：[1]桂林电子科技大学信息与通信学院,广西桂林541004

出　　处：《桂林电子科技大学学报》2018年第5期422-426,共5页Journal of Guilin University of Electronic Technology

基　　金：国家自然科学基金(61662012)

摘　　要：针对传统英语作文跑题检测方法需多篇参考范文进行对比的局限性,提出一种基于文档分布式表示的英语作文跑题分析模型。该模型将词的分布式表示方法与文档的分布式表示方法相结合,从而充分利用作文标题的语义信息对待测作文进行跑题分析,并在真实的数据集中进行实验。实验结果表明,该方法仅需作文标题就可以有效判断英语作文是否跑题,并且具有较高的准确率和较好的实用性。Traditional off-topic English essay detection method required several model essays for comparison,for these kinds of cases,an English off-topic analysis model based on distributed representation of documents is proposed.Distributed representation of words is combined with distributed representation of documents so as to make full use of the essay titles’ semantic information.Via experimenting with real data sets,the results show that the proposed model only needs essay titles to realize the function of judging whether English essay is off topic or not and has high accuracy and good practicability.

关 键 词：英语作文 跑题检测 文档分布式表示 文本相似度 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]