关键词: 字符层马尔科夫模型 语种识别 机器翻译 多语种 马尔科夫模型 识别 字符 Markov 训练算法 预处理过程

Abstract: Language identification is a necessary pre-process in machine translation and other muhi-language applications, but no experiments hase yet been reported on double-byte encoded languages, such as Chinese and Japanese. An efficient EM based training algori

Key words: Character based markov models, Language identification, Machine translation