Chinese Base Noun Phrase Based on Multi-Class Support Vector Machines and Rules of Post-Processing

In the paper, Chinese base noun phrase chunking is considered as a classification problem, and the paper proposes an approach, combines SVM-based method and rules of post-processing method, to distinguish Chinese base noun phrase. But the paper introduce threshold in multi-class SVM algorithm and study the usefulness of threshold for Chinese base noun phrase chunking, complete analyses of the result in this paper, then according to the special structures of Chinese base noun phrase, customize some appropriate rules to process the result. From overall experiments, the method achieves a higher accuracy in the final results.