图书情报工作 ›› 2013, Vol. 57 ›› Issue (03): 120-124.DOI: 10.7536/j.issn.0252-3116.2013.03.022

• 知识组织 • 上一篇    下一篇

基于规则的纪传体古代汉语文献姓名识别

皇甫晶1, 王凌云2   

  1. 1. 陕西科技大学图书馆;
    2. 广联达软件股份有限公司
  • 收稿日期:2012-11-05 修回日期:2012-11-28 出版日期:2013-02-05 发布日期:2013-02-05
  • 作者简介:皇甫晶, 陕西科技大学图书馆助理馆员,硕士,E-mail:huangfu774@163.com;王凌云,广联达软件股份有限公司开发工程师,硕士。

Rule-based Chinese Person Names Identification in Ancient Chinese Literature of Annals-Biography (Jizhuan) Style

Huang Fujing1, Wang Lingyun2   

  1. 1. Shaanxi University of Science and Technology Library, Xi'an 710021;
    2. Glodon Software Company Limited, Beijing 100193
  • Received:2012-11-05 Revised:2012-11-28 Online:2013-02-05 Published:2013-02-05

摘要:

设计一个可以自动识别古代汉语文献中姓名的模型系统,对纪传体古代汉语文献中的姓名识别作了实验和探索。以晋陈寿的《三国志·蜀书》十五卷为实验文本,对系统的识别效果进行测试,识别结果为召回率75.4%,准确率91.9%。实验证明,基于规则的方法对于识别纪传体古代汉语文献中的姓名是可行的。

关键词: 命名实体识别, 中文姓名识别, 古代汉语文献, 纪传体, 基于规则

Abstract:

This paper designs a model system to automatically identify person names in ancient Chinese literature of annals-biography (Jizhuan) style and makes some explorations. This model system is tested by the experimental text which is composed of 15 volumes of Book Shu of Annals of the Three Kingdoms written by Chen Shou of Jin Dynasty. The recognition result is 75.4% as the recall ratio and 91.9% as the precision ratio. The result shows that the ruled-based method is feasible to identify person names in ancient Chinese literature of annals-biography (Jizhuan) style.

Key words: named entity recognition, Chinese person names identification, ancient Chinese literature, annals-biography (Jizhuan) style, ruled-based

中图分类号: