Combining Constituent and Dependency Syntactic Views for Chinese Semantic Role Labeling

摘要

This paper presents a novel feature-based semantic role labeling (SRL) method which uses both constituent and dependency syntactic views. Com-paring to the traditional SRL method relying on only one syntactic view, the method has a much richer set of syn-tactic features. First we select several important constituent-based and de-pendency- based features from existing studies as basic features. Then, we pro-pose a statistical method to select dis-criminative combined features which are composed by the basic features. SRL is achieved by using the SVM classifier with both the basic features and the combined features. Experimen-tal results on Chinese Proposition Bank (CPB) show that the method outper-forms the traditional constituent-based or dependency-based SRL methods.