本文探讨了基于字字同现、词性二元接续和语义二元接续的中文文本的自动查错原理和查错算法;给出了字词接续判断模型,并讨论了与接续判断模型相关的查错知识库的构造方法。通过对实验结果的分析和评测,证明本文所述方法是可行的。
Abstract
Automatic error detecting principle and algorithm of Chinese texts based on character-character co-occurrence , POS bi-neighborship and semantic bi-neighborship are discussed in this article. The models of judging character and word neighborship are presented ,and the method of constructing error detecting knowledge bases which is related to these models is introduced. According to the analysis and estimation for experiment results , the method given in this paper is workable.
关键词
中文文本自动校对 /
自动查错 /
二元接续关系
{{custom_keyword}} /
Key words
Chinese text automatic proof read /
automatic error-detecting /
bi-neighborship
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 慕勇,孙才,罗振声. 汉语文本自动查错与确认纠错系统的研究. 见:计算语言学进展与应用,北京:清华大学出版社,1995
[2] 邱超捷,宋柔等. 大规模语料库中词语接续对的统计与分析. 见:语言工程. 北京:清华大学出版社,1997
[3] 张仰森,丁冰青. 中文文本自动校对技术现状及展望. 中文信息学报,1998 ,12 (3)
[4] 于勐,姚天顺. 一种混合的中文文本校对方法. 中文信息学报,1998 ,12 (2)
[5] 刘开瑛. 中文文本自动分词和标注. 北京:商务印书馆,2000年
[6] 张永奎等,基于义类组合信息的义类排歧方法研究. 情报学报,1996.12 ,增刊
[7] James L Peterson. Computer Programs for Detecting and Correcting Spelling Errors. Communication of the ACM 80. 12
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
山西省自然科学基金(981031)
{{custom_fund}}
PDF(350 KB)
