摘要
句子处理在自然语言信息处理中处于关键位置,它是为篇章处理和语义分析打基础。该文描述蒙古语短语树库,实现了蒙古语句子的自动分析,目前自动分析处于浅层处理。同时对分析器进行了测评,得出不同长度的句子准确率会不同,而分析器准确率也会因句子长度增加而下降的结论。因此,还仔细研究了自动分析错误的句子,归纳出自动分析中遇到的难点,为进一步改善分析器分析能力提供理论依据。
Sentence processing plays an important role in natural language information Processing. It is thefoundation of text processing and semantic analysis. This paper describes the characteristics of Mongolian phrasetree,the realization of the automatic analysis of Mongolian sentences,automatic analysis is in shallow processing.At the same time,the parser was tested and evaluated. And the accuracy of the sentences with different lengths isdifferent,the accuracy of the parser will be decreased due to the increase of the length of the sentence. Therefore,we have also studied the sentences that automatically analyzed the errors,summed up the difficulties encountered in the automatic analysis,and provided the theoretical basis for further improving the analytical ability of the parser.
作者
乌兰
达胡白乙拉
Wu lan;Dabhurbayar(Department of Mongolian Language and Literature＇ Minzu University of China,Beijng 100081;College of Mongolian Studies,Inner Mongolia University,Hohhot 010021,China)
出处
《内蒙古民族大学学报（社会科学版）》
2018年第2期17-21,共5页
Journal of Inner Mongolia Minzu University：Social Sciences
关键词
蒙古语
短语结构
句法自动分析
Mongolian
Phrase structure
Syntactic automatic parsing
