计算机自动分析量词短语的方法及规则

The Methods and rules of automatic analysis of Chinese classifiers on computer

本文对现代汉语量词进行了分析和总结,从组合关系的角度分析量词在句段中的地位,根据量词与其它词的搭配关系总结出了有关分析量词短语的语法规则。在量词中,有一些量词本身之间是同音词,由于它们的语法功能相同,所以单纯用语法上下文已无法区分它们了,必须求助于语义上下文及不同量词与名词的搭配不同来分析它们,本文利用量词与不同语义的名词的搭配关系来区分同音量词,这些方法及规则可应用于同音词的自动识别,汉外机器翻译等方面,我们已在微机上用C语言实现了同音量词自动识别系统。

This paper is inteded to explore ways of applying grammar, semantics and context, etc. to the automtic analysis of Chinese Classifiers on computerolto analyses the collocation between classifiers and other parts of speech from the view point of syn?tagmatic relations and summarizes the grammatic rules for analysis of Chinese Classifiers as follows.l. Determinative + (other words) + Classifier + (other word) + (Noun)2. Verb + Classifier + (other word) + (Noun)3. Classifier + classifier + (other word) + (Noun)Among all Classifiers, however, still exist a number homonyms, or homonyms with same parts of speech, they cannot be simply differentiated by grammatical con-etxt alone, but only by sematic context. In this case, a commonly-used collocations of classifiers and nouns should be found out. In this paper, each group of homony-mous classifiers is analysed so as to study how to differentiate these homonyms based on the collocation between homonymous classifiers and nouns. The grammatical and semantic rules suggested in the paper can be used in automatical differentiation of Chinese homonyms and machine translation from Chinese into foreign languages.The automatic reaognition system of homonymous classifiers has been implemented with C language on miarocomputer by us.