摘要
汉语的歧义分布在语言的不同层面上,从词形变化到句子结构都存在歧义.针对汉英机器翻译不同阶段遇到的歧义问题,采用了隐马尔柯夫模型和贝叶斯分类法来进行排歧.实验表明:基于统计的多步消歧策略在汉英机器翻译系统中具有较高的排歧准确率.
Distributed in every layer,the Chinese ambiguities exist in both metaplasm and sentence structure. Based on the ambiguity problems we meet in every stage of Chinese - English machine translation, we use the methods based on the model of hidden Markov and Bayes to disambiguate. The experimental results show that the methods of statistical disambiguation can achieve a high disambiguation accuracy rate in Chinese - English machine translation system.
出处
《西南民族大学学报（自然科学版）》
CAS
2006年第1期191-194,共4页
Journal of Southwest Minzu University(Natural Science Edition)
基金
四川省科技厅攻关基金资助(项目编号:05SG022-016)
关键词
消歧
机器翻译
切分歧义
标注歧义
多义词歧义
disambiguation
word sense disambiguation
tag disambiguation
word sense disambiguation
