Computer Engineering and Applications ›› 2015, Vol. 51 ›› Issue (11): 6-11.

Previous Articles     Next Articles

Automatic identification of verb-preposition multi-category words for Chinese-English patent machine translation

LI Hongzheng, ZHU Yun, JIN Yaohong   

  1. 1.Institute of Chinese Information Processing, Beijing Normal University, Beijing 100875, China
    2.China Patent Information Center-Beijing Normal University Joint Laboratory of Machine Translation, Beijing 100875, China
  • Online:2015-06-01 Published:2015-06-12


李洪政,朱  筠,晋耀红   

  1. 1.北京师范大学 中文信息处理研究所,北京 100875
    2.中国专利信息中心-北京师范大学机器翻译联合实验室,北京 100875

Abstract: Multi-category words are widely distributed in Chinese patent documents. A rule-based method is presented to identify verb and preposition multi-category words in Chinese-English patent machine translation. Based on the principles of boundary perception, and grammatical and semantic information of multi-category words, as well as the context information, serials of disambiguation and identification strategies are designed, which are described in formal rules. Related experiments show the method is efficient to identify verb and preposition multi-category words, and is helpful to improve final translation results.

Key words: verb, preposition, multi-category words, identification, rule, machine translation

摘要: 兼类词在汉语专利语料中分布普遍。面向汉英专利机器翻译提出了一种基于规则的动-介兼类词识别方法。根据边界感知原则和兼类词的句法语义属性以及周围的语境信息,设计了一系列兼类排歧规则,同时分别提出了兼类词识别为动词或介词的策略,并以形式化的规则加以描述。相关实验测试表明提出的方法可以有效地识别动-介兼类词语,对改善翻译系统的翻译结果也有帮助。

关键词: 动词, 介词, 兼类词, 识别, 规则, 机器翻译