论文部分内容阅读
本文将通过传统藏文文法的语法规则,主要研究藏文文本中大量藏文不自由虚词的识别算法,同时建立了藏文不自由虚词的消岐规则库。使计算机快速地识别并消除藏文句子中不自由虚词的歧义问题。提高藏文自动分词的准确率。使句子的生产、句法分析、八格识别和机器翻译等研究的基础更扎实。
In this paper, through the grammatical rules of traditional Tibetan grammar, this paper mainly focuses on the recognition algorithms of a large number of non-free Tibetan function words in Tibetan texts. At the same time, a disambiguation rule base of Tibetan non-free function words is established. So that the computer quickly identify and eliminate ambiguous words in Tibetan sentences are not free words. Improve the accuracy of Tibetan automatic word segmentation. Make the basis of the research of sentence production, syntactic analysis, eight-grid identification and machine translation more solid.