基于音码相似度的拼音模糊查询算法

An Arithmetic of Speech Fuzzy Query Based on Spelling Similarity

论述了拼音模糊检索技术在信息管理和网络信息搜索系统中的必要性,描述了基于音码相似度的语言模糊查询算法及实现同音字和近音字检索算法,在中文信息检索中有很好的应用价值。并结合实例,在获得同音字数据库基础上,提出了基于音码相似度阈值的模糊查询算法,给出了通过拼音数据库实现中文全拼和首字母简拼检索数据库字段的实现方案,从查全率和查准率两个方面对算法的检索效果进行了评价,同时分析了音码相似度阈值对查全率和查准率的影响。

This paper discusses the necessary of applying speech fuzzy query technique to information management system and Web information search system, describes the speech fuzzy query arithmetic and the method of realizing homophone or similar sound words query, this technique plays all-right role in information retrieval, and with examples, on the bases of obtaining hom- ophone words database, gives the way of achieving full spelling or the first character of Chinese words, and further more, by the rate of full query and exact query, evaluates the query effect of this arithmetic, at the same time, analyses the influence of spelling similarity clique on the rate of full query and exact query.