一种试题智能提取与批量导入方法  

A method of intelligent extraction and batch import of test questions

作　　者：王东 林宏 WANG Dong;LIN Hong(Guizhou Education University,Guiyang 550018,Guizhou China)

机构地区：[1]贵州师范学院数学与大数据学院,贵州贵阳550018

出　　处：《贵阳学院学报（自然科学版）》2020年第1期87-92,共6页Journal of Guiyang University：Natural Sciences

基　　金：贵州省科学技术基金计划项目(项目编号:黔科合基础[2016]1115);贵州省教育厅科技拔尖人才支持项目(项目编号:黔教KY字[2017]081);贵州师范学院2018年度校级重点课题(项目编号:2018ZD005)。

摘　　要：试题批量导入方法是试题库系统的重要研究内容。针对已有方法存在操作复杂、信息丢失、效率低下等问题,提出了一种试题智能提取与批量导入方法。首先设置简单易用的标识规则,制定导入文档模板,然后设计了试题自动识别与分割算法,逐题获得试题题干和答案,最后使用基于向量空间模型的试题相似度检测方法对试题重复性进行识别,再将试题信息存入数据库。实验表明:与单题录入方法相比,本方法的时间效率优势明显,适合大批量的试题录入。Batch import method of test questions is an important part of the research of question bank system.Aiming at the problems of complicated operation,information loss and low efficiency of existing methods,an intelligent extraction and batch import methodof test questions is proposed.Firstly,we set up simple and easy-to-use identification rules and formulate import document template.Then an automatic recognition and segmentation algorithm is designed to obtain the stem and answer of each question.Finally,the similarity detection method based on vector space model is used to identify the repeatability of the test questions,and then the information of the test questions is stored in the database.Experiments show that,compared with one-by-one entry method,the batch input method has obvious advantages in time efficiency and is suitable for large-scale test entry.

关 键 词：试题批量导入 试题录入 试题库系统 相似度检测 向量空间模型 

分 类 号：TP311.52[自动化与计算机技术—计算机软件与理论]