基于反馈规则学习的医学文献主题自动标引方法

Feedback Rule-based Learning Medical Literature Subject Automatic Indexing

就中医药文献的自动标引研究,介绍了一种基于规则学习的主题自动标引方法? 与以往基于词频统计和加权的自动标引方法,基于反馈的规则学习的方法能有效提取文献的 副主题词,并进行主/副题词组配,具有很好的扩展性和适应性,基于此方法开发的系统在 大量中医药文献中作了实验,获得了很好的标引结果?

This paper presents a feedback rule-based learning method of automat ic indexing (FRBLAI) for medical literature. Compared with the term frequency an d weight-based approaches, FRBLAI can make efficient extraction of subheadings a nd its combination of corresponding headings. Also FRBLAI avoids the difficulty of segmentation in Chinese text. And finally it gives an automatic indexing expe riment in medical literature.