论文类型：会议论文
收录刊物：Scopus、CPCI-S、EI
卷号：1
页面范围：731-734
摘要：This paper presents a method of Chinese text chunking based on editing support vector machine (ESVM) and K nearest neighbors (KNN). The word itself part-of-speech (POS) tag, syllable and context information are extracted as the features of the vectors. The experimental results show that this model is efficient for Chinese text chunking. The hybrid machine learning model based on ESVM and KNN can achieve better results than SVM. The recall, precision and F-measure are up to 84.11%, 83.01% and 83.56% respectively in open test. And the combined ESVM-KNN model can be generalized to the fields of machine learning with unbalanced class distribution.