摘要
本文是现代书面藏语不同类型句法单位变体形式的机器自动识别研究,包括词形 变化的单位分类,抽取变体形式的信息和对文本进行句法语义功能的标注处理方法。
This paper discusses how to build up machine identification rules for differentiating the heterographs of Tibetan words and markers on the basis of affixing-rules database. The Tibetan affixing-rules database contains more than ten types of words, which change forms in texts according to the pre-syllabic ending forms. So the paper first describes the types of heterographic variants and their conditions, then advances a method to extract information of variants and goes on tagging and processing all the heterographs in the text.
出处
《民族语文》
CSSCI
北大核心
2004年第6期50-54,共5页
Minority Languages of China
基金
国家自然科学基金资助(项目批准号为60173024)。
