摘要
主题提取是文本处理的一项重要工作。本文首先分析了主题抽取中加权方法形成时的一些定量问题 ,然后提出了主题相关词一种非线性加权处理方法 ,对比实验结果显示它不仅是一种比较稳健的方法 ,而且能在一定程度上提高主题提取的正确率。
Subject distillation is an important task in text information processing. In this paper, we analyse some quantitative problems in term weighting when extracting subject from texts, then put forward a non linear term weighting method in detail. Our experiments show that it is not only a robust, but also a useful method to rise the precision of subject distillation.
出处
《情报学报》
CSSCI
北大核心
2000年第6期650-653,共4页
Journal of the China Society for Scientific and Technical Information
