Chinese Semantic Role Labeling Based on Genetic Algorithm
Ning Mao, Shao Yanqiu, Liang Chunxia · 2012 IEEE/WIC/ACM International Conferences on Web Intelligence and Intelligent Agent Technology · 2012
To find the effective features from many syntactic features could improve the efficiency and accuracy of the semantic role labeling system. Based on the original semantic role labeling system, the genetic algorithm is used to optimize those syntactic features. From the experiment results, it could be concluded that the syntactic feature selection based on genetic algorithm is efficient. The system uses fewer features, but achieves almost the same F value as the whole extended model.