自动抽词与自动分词  

Automatic Extraction and Separation of Terms

作　　者：张琪玉[1] 

机构地区：[1]南京政治学院上海分院信息管理系,教授200433

出　　处：《图书馆杂志》2002年第3期13-14,共2页Library Journal

摘　　要：自动抽词与自动分词既有紧密联系又有重大差别。自动抽词标引除编制题内关键词索引外,至今没有突破性进展,主要是检准率太低。本文指出自动抽词标引研究注重在各种更有效的算法的寻找,而忽视对相关问题的深入、系统的研究,是其进展缓慢的重要原因之一,并具体列举了一些相关问题。Automatic extraction of terms is closely linked with but different from automatic separation of terms. Except for keyword - in - context index, there is no breakthrough made in automatic derived indexing today due to low precision ratio of retrieval. The paper points out that the reason why slow progress has been made is that people has paid much attention to various more effective ways of algorithm but ignored in - depth and systematic research on the relative issues listed when making research on automatic derived indexing.

关 键 词：自动抽词 自动分词 文献标引 文献检索 人机结合 

分 类 号：G254[文化科学—图书馆学]