Author Login Chief Editor Login Reviewer Login Editor Login Remote Office

Computer Engineering ›› 2007, Vol. 33 ›› Issue (09): 31-33.

• Degree Paper • Previous Articles     Next Articles

Identification of Chinese Names Based on Maximum Entropy Model

JIA Ning1,2, ZHANG Quan2   

  1. (1. Graduate School, Chinese Academy of Sciences, Beijing 100039; 2. Institute of Acoustics, Chinese Academy of Sciences, Beijing 100080)
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-05-05 Published:2007-05-05

基于最大熵模型的中文姓名识别

贾 宁1,2,张 全2   

  1. (1. 中国科学院研究生院,北京 100039;2. 中国科学院声学研究所,北京 100080)

Abstract: Maximum entropy model can be used to found probability model of many natural language problem. This paper presents an approach for Chinese name recognition based on statistic model. It founds the model with maximum entropy principle, and calculates name probability by the model. It gets good recalling rates and precision by open test on actual corpus. The method is practicable.

Key words: Chinese name recognition, Maximum entropy, Feature function

摘要: 最大熵模型可以对非常广泛的自然语言现象建立概率模型,该文提出了一种使用统计模型的中文姓名识别方法,采用最大熵原则建立统计模型,并运用该模型计算姓名概率。系统采用真实语料进行开放测试的结果表明,在保证一定准确率的情况下,系统召回率可达90%以上。

关键词: 中文姓名识别, 最大熵, 特征函数

CLC Number: