基于概念树扩展的中文文本检索研究  

Research on Chinese text retrieval based on expansion of concept tree

作　　者：张映海[1] 

机构地区：[1]武警广州指挥学院计算机教研室,广州510440

出　　处：《计算机工程与应用》2008年第26期154-157,共4页Computer Engineering and Applications

基　　金：国家自然科学基金No.60173060~~

摘　　要：分析了概念在语义层次上的扩展,将概念树中的父子概念关系用词语的相似度进行量化。提出了检索概念权重计算的两种方法和一种基于检索概念的文本概念权重计算方法,并将这些方法用于中文文本检索,因此,构建了基于概念树扩展的两个文本检索模型。实验显示,这两个检索模型的精确率与关键词检索模型保持基本一致,召回率却得到较大提高。The expansion of semantic levels of concept is analyzed,the relationship between father and son concept in concept tree is translated by the similarity of words.Two methods,which are based on computing retrieval concept weight,and one method,which is based on computing text concept weight based on retrieval concept,are proposed.These methods are used in Chinese text retrieval.So,two text retrieval models are constructed based on concept tree expansion.Experiments show that the precision of the two retrieval models can be in line with keyword retrieval model,but the recall rate is improved greatly.

关 键 词：中文文本检索 概念树 概念权重 检索模型 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]