藏文词性自动标注是藏文信息处理后续句法分析、语义分析及篇章分析必不可少的基础工作。词性歧义问题的处理是藏文词性自动标注的关键所在,也是藏文信息处理的难点问题。对藏文词性标注中词性歧义问题进行了分析研究,提出了符合藏丈语法规则实用于藏文词性标注的解决词性排岐方法。实验证明:该处理方法在藏文词性自动标注中对词性排岐方面有较好的效果,使藏文词性标注正确率有了一定的提高。
Tibetan language Part-Of-Speech(POS) tagging is the subsequent parsing of Tibetan language information processing. POS tagging is an essential foundation work for semantic analysis and text analysis. POS ambiguity problem solving is the key to Tibetan POS tagging, is also one of the difficulties in the Tibetan automatic POS tagging. This paper analyzes and studies POS ambiguity problem in the Tibetan POS tagging, and puts forward a method of solving POS ambiguity problem suitable for Tibetan grammar rules. Experiments prove that this method on speech disambiguation in the Tibetan POS tagging has achieved better results and has definitely increased the accuracy of the Tibetan POS tagging.