自学习结合部分句法分析的汉语词性标注

A Self-Learning Part of Speech Tagging Integrated with Partial Syntactic Analysis

针对词性标注中单独使用概率方法或规则方法的缺陷 ,将概率方法和规则方法有机地结合起来 ,并引入了部分句法分析排除切分歧义和标注歧义 .利用机器学习得到的规则对输出结果进行修正 。

Considering the disadvantages of the statistical approach and rule approach in the part of speech tagging, a statistical approach is integrated efficiently with the rule approach. This paper excludes the ambiguity of segmentation and part of speech tagging by the partial syntactic analysis and amends the results with the rules acquired by machine learning. It completed a self learning part of speech tagging integrated with the partial syntactic analysis.