摘要
根据试题具有一般文本的特点,提出把计算机分类技术应用于试题分类。借鉴文本分类的关键技术,成功创建了一个基于向量空间模型的试题分类系统。把全国专业技术人员计算机应用能力考试的"PowerPoint2003中文演示文稿"模块题库作为试题语料,进行试题分类实验,结果表明了该试题分类系统的可靠性。同时探讨了如何利用试题分类系统对题库进行质量控制。
According that question has the features of general text, The idea of applying computer categorization technology to question categorization is put forward. A lesson is drawn the key technology of text categorization and a question categorization system based on vector space model is set up. The PowerPoint 2003, Chinese demonstration manuscripts module, which is the national professional and technical personnel computer skill test, is used as question corpus in the experiment of question categorization. Through the experiment, the reliability of question categorization system is proved. How to use the question categorization system to control the quality of the question library is discussed.
出处
《计算机工程与设计》
CSCD
北大核心
2008年第12期3227-3229,3233,共4页
Computer Engineering and Design
基金
广东省科技攻关基金项目(2005B10101033)
广州市科技攻关基金项目(2006Z3-D3051)
关键词
试题分类
文本分类
向量空间模型
相似度
质量控制
question categorization
text categorization
vector space model
similarity
quality control
