计算机应用与软件2017，Vol.34Issue(10)：85-90,142,7.DOI:10.3969/j.issn.1000-386x.2017.10.014
一个基于词语搭配的英文文本检索软件的实现
AN IMPLEMENTATION OF ENGLISH TEXT RETRIEVAL SOFTWARE BASED ON WORD COLLOCATION
摘要
Abstract
Word collocation is an important subject in the study of English linguistics.In recent years,it tends to focus on data validation and quantitative research.This paper discusses the key technology of ColloStu,an English text retrieval software based on collocation research.The software designs a wildcard matching algorithm that uses the DFA to speed up the matching speed by compressing the number of its states.It can identify the sentence terminator in the cooccurrence context in order to retrieve the collocations more effectively.We have improved the Z score algorithm of collocation calculation.We use Z score,T score and MI value to compute collocation intensity from multiple angles to make the calculation more accurate.Experiments show that,compared with the mainstream search software,ColloStu addition to adding the collocation calculation function,its word statistics and collocation word search is more accurate.关键词
文本检索/词语搭配/通配符匹配/确定有限自动机/搭配力计算Key words
Text retrieval/Word collocation/Wildcard matching/DFA/Collocation calculation分类
信息技术与安全科学引用本文复制引用
乔艳梅,杨进才,刘应亮..一个基于词语搭配的英文文本检索软件的实现[J].计算机应用与软件,2017,34(10):85-90,142,7.基金项目
国家社会科学基金项目(14BYY093) （14BYY093）
国家自然科学基金项目(31371275). （31371275）
