基于Lucene的语段模糊匹配中文检索系统设计与实现

A Chinese Section Fuzzy Matching Information Retrieval System Based on Lucene

为提高图书馆中文信息检索的精确度和有效性,设计了基于Lucene的语段模糊匹配中文检索系统。其采用了自然语言处理中的词语切分技术,使输入条件可以直接通过自然语言的方式提交,同时针对语段匹配的实际问题情境,设计了一种新的结果有效性判别模型,提高了检索结果相似度的科学性和准确性。经过多次实验结果的统计,搜索结果有效性可提高12%。

For the purpose of improving the veracity and validity of Chinese information retrieval in library, the article designs the Chinese section fuzzy matching information retrieval system which is based on Lucene uses natural language split technology, and input can be directly submitted by natural language. Meanwhile, the paper designs a new model of validity estimation for the practical instance of section matching, improves the scientificity and veracity of similarity calculation of result aggregation. Several statistics indicate the validity of searching result is improved about twelve percents.