神经网络与规则相结合的词类标注方法

词类标注是自然语言理解中的的一个关键问题，是句法分析和语义分析的基础．本文分析了汉语的多类词现象．提出了在基于神经网络的词类标注方法的基础上，根据汉语的语法结构增加了规则的排歧处理阶段，增加了上下文词类信息的处理能力．实验表明：在原有神经网络方法的基础上，利用规则的方法进行修正，可以提高词性标注的正确率．

Hybrid neuro and rule based part of speech taggers

A hybrid system for tagging part of speech consists of a neuro tagger and a rule based corrector. Part of speech tagging is one of the crucial stages in the natural language process.During the tagging process,the neuro tagger is an initial states annotator that uses different lengths of contexts based on longest context priority.Then the rule based process corrects the result for the shortcomings of the neuro tagger.