Automatic Chinese Text Classification Based on NSVMDT-KNN

Author

Xu, QiNan ; Liu, Zhijng

Author_Institution

Sch. of Comput. Sci. & Technol., Xidian Univ., Xian

Volume

2

fYear

2008

fDate

18-20 Oct. 2008

Firstpage

410

Lastpage

414

Abstract

According to this paper, a novel approach based on non-linear support vetor machine decision tree (NSVMDT) and K nearest neighbors (KNN) is proposed towards Chinese text categorization. To begin with, SVM is extended to non-linear SVM by using kernel functions. And then the method of NSVMDT is presented based on traditional SVM decision tree. Furthermore, the KNN is combined with NSVMDT to solve the problem of the categorization of unbalanced Chinese texts sets. According to this method, experimental results have shown that the hybrid method based on NSVMDT and KNN could achieve better results than traditional SVM method for Chinese text categorization.

Keywords

decision trees; support vector machines; text analysis; K nearest neighbors; NSVMDT-KNN; automatic Chinese text classification; kernel functions; nonlinear support vector machine decision tree; Decision trees; Entropy; Feature extraction; Fuzzy systems; Mutual information; Statistics; Support vector machine classification; Support vector machines; Text categorization; Text processing; Chinese text categorization; K nearest neighbors (KNN); kernel functions; non-linear support vetor machine decision tree (NSVMDT);

fLanguage

English

Publisher

ieee

Conference_Titel

Fuzzy Systems and Knowledge Discovery, 2008. FSKD '08. Fifth International Conference on

Conference_Location

Shandong

Print_ISBN

978-0-7695-3305-6

Type

conf

DOI

10.1109/FSKD.2008.289

Filename

4666149