藏文信息处理技术随着对语言文字信息处理研究工作的不断深入,逐步从字信息处理转向语言信息处理。与汉语、日语等语种的信息处理一样,藏文自动分词是藏文信息处理中一项必不可缺的工作。但是目前藏文自动分词系统的研制在国内仍是空白。本文从四个方面详细报告了书面藏文自动分词系统的具体实现形式,以及在实现过程中出现的歧义类型及如何进行有效消除歧义方法的探讨,通过实验结果和实例分析来进行本论文的研究。 With the deepening of research work on the processing of language information, Tibetan information processing technology has gradually shifted from word information processing to language information processing. As with the processing of information in languages ​​such as Chinese and Japanese, automatic word segmentation in Tibetan is an indispensable task in Tibetan information processing. However, the development of Tibetan automatic word segmentation system is still blank in the country. This article reports in detail four aspects of the concrete realization forms of written Tibetan automatic word segmentation system, as well as the types of ambiguities that arise in the process of realization and how to effectively disambiguate the disambiguating methods. The research of this dissertation is carried out through experimental results and case studies .