基于关键词匹配技术的相似试题检测方法研究

The Application of Keyword Matching Technology in Similar Question Detection

在高校题库内容重复率是评价题库建设质量的一个重要指标,为了快速找到题库中重复题或相似度很高的试题,本文主要研究了基于关键词匹配技术的相似试题检测方法 (KMQT)。该方法首先使用基于词典的分词方法将题干内容分解出若干关键词,然后利用KMP算法对关键词进行字符串匹配,最后按照匹配成功的关键词的个数对结果集进行排序。通过在题库系统中的使用,充分验证了此方法的可行性,并达到了很好的效果。

Test content repetition rate in colleges and universities is an important indicator to evaluate the quality of the database construction. In order to locate same or similar items in question bank, this paper mainly studied the similar test detection method based on key word matching technology （KM QT）. This method is first used to dry the question word segmentation method hased on dictionary content into a number of key words, and then by KMP algorithm for string matching keywords, finally sort out the result by matching the number of keywords. Experiments prove to be practicable and effective.