ISSN 0439-755X
CN 11-1911/B
主办:中国心理学会
   中国科学院心理研究所
出版:科学出版社

心理学报 ›› 1988, Vol. 20 ›› Issue (3): 5-13.

• • 上一篇    下一篇

机器理解古汉语的实验和若干问题

陈永明;李家治;蔡山   

  1. 中国科学院心理研究所 ,中国科学院心理研究所 ,中国科学院心理研究所
  • 出版日期:1988-09-25 发布日期:1988-09-25
  • 通讯作者: 陈永明

A STUDY ON COMPUTERIZED INTERPRETATION OF ANCIENT CHINESE AND RELATED PROBLEMS

Chen Yong-ming;Li Jia-zhi;Cai Shan Institute of Psychology, Academia Sinica, Beijing   

  • Published:1988-09-25 Online:1988-09-25

摘要: 本文介绍了机器理解古汉语的一个实验系统(ACLUS)。该系统由一个知识库和三个子程序组成。知识库包含有一部机器词典和若干集用来加工句子的规则。三个子程序是:句法分析子程序、翻译子程序和问答子程序,它们分别执行三种不同的功能。本文还讨论了有关机器理解古汉语的若干重要问题。

关键词: NULL

Abstract: This paper describes an ancient Chinese language understanding system(ACLUS) in processing ten proverbial stories written in ancient Chinese. The system consists of a knowledge base and three subprograms. Theknowledge base includes two parts, one is a dictionary; the other consistsof a few sets of rules for processing sentences. The words in the dictio-nary are annotated with their lexical meanings and syntatic and semanticfeatures, and also with information on semantic match between some verbsand nouns. The syntatic analysis subsystem makes sentence parsing and providesspecial treatment of function words (FW), The translation subsystem trans-lates stories from ancient into modern Chinese, The question-and-answersubsystem behaves like a dialogue between teacher and student about themeanings of words and the meanings or grammatical construction of somesentences. The paper discusses the problems about case analysis, the tense of theverb and the addition or deletion of words in a sentence in ancient Chineseso as to facilitate interpretation and translation. The authors point outthat there is a remarkable difference between Chinese and Western langu-ages, and that the Chinese have their own way of understanding theirmother tongue. To identify the case of a noun or the tense of a verb, theChinese do not depend on the morphological signs of the word, but on abroad understanding of the context.

Key words: NULL