网络环境中汉语叙词表的自动构建研究  

Research on Automatic Construction of Chinese Thesaurus

作　　者：杜慧平[1] 侯汉清[2] 

机构地区：[1]上海师范大学图书馆,上海200234 [2]南京农业大学信息管理系,南京210095

出　　处：《情报学报》2008年第6期863-869,共7页Journal of the China Society for Scientific and Technical Information

摘　　要：为解决网络信息检索效率低的问题，需要把叙词表等控制机制引入到网络检索系统中，但传统词表编制方式已经不能满足网络信息检索的需要。本文制定了一种汉语叙词表自动构建方案，以财税领域为例，采用模式识别、词聚类、同现分析等自然语言处理技术实现自动识别词间等同关系、等级关系和相关关系，从而构建一部财税领域叙词表，最后对所构建的叙词表进行测评。相比传统叙词表编制方法，自动构建叙词表能降低编表专家的智力负担，而且编表时间短，费用低，时效性强，符合普通用户的检索习惯。但对词间关系的识别不如人工编表时精当可靠，所以需要人工辅助判定。Thesaurus is needed to resolve the problem of low information retrieval efficiency. Traditional methods of thesaurus construction can＇t meet the need of high retrieval efficiency, so this article approaches an automatic approach, natural language processing technologies such as pattern-recognition, word clustering, co-occurrence analysis are used to construction a Chinese taxation thesaurus, and the thesaurus generated is tested at last. Compared with the traditional ways, this method can relieve expert＇s mental burden, improve information retrieval efficiency with lower cost and shorter time-consuming, but the association between words are not as strong in semantics as those in handcrafted ones, so a little laborious human involvement is also needed.

关 键 词：叙词表自动构建 网络信息检索 信息组织 

分 类 号：G254.24[文化科学—图书馆学] G250.73