摘要
情报检索的目的是为用户服务 ,因而标引词的提取应以其与文献主题内容相关程度为标准。文章基于原有的统计分析标引法 ,对其权值设计予以重新考虑 ,并与文献词频统计相结合 ,使分词与标引相统一 ,标引词更好地反映文献主题概念 ,提高检索效率。
Because the aim of information retrieval is to serve the users,the correlated degree of indexing words and the subject of the document should be the criteria of drawing them.Based on the original statistical analysis method of indexing,the author reconsiders its words weight design and connects it with words frequency statistic,unifying the work of dividing words and indexing,so that the main idea of the document can be better reflected and the retrieval efficiency can be improved.
出处
《情报学报》
CSSCI
北大核心
2000年第4期333-337,共5页
Journal of the China Society for Scientific and Technical Information
关键词
自动标引
词频统计
权值
统计分析
automatic indexing,words frequency statistic,words weight.