Improved Algorithm for Text Classification Based on TSVM

Author

Teng, Guifa ; Liu, Yihong ; Ma, Jianbin ; Wang, Fang ; Yao, Huiting

Author_Institution

Sch. of Inf. Sci. & Technol., Agric. Univ. of Hebei

Volume

2

fYear

2006

fDate

Aug. 30 2006-Sept. 1 2006

Firstpage

55

Lastpage

58

Abstract

TSVM (transductive support vector machines) tries to minimize misclassification of these particular examples based on a particular test set. It is more practical and performs well on classification. In this paper, a progressive SVM is introduced briefly and an improved algorithm for text classification named double transductive inference algorithm based on TSVM is presented in detail. The experimental results on e-mail classification show that this improved algorithm is effective on a mixed training set of a small number of unlabeled examples and a large number of labeled examples

Keywords

classification; learning (artificial intelligence); support vector machines; text analysis; TSVM; double transductive inference algorithm; e-mail classification; text classification; transductive support vector machines; Classification algorithms; Electronic mail; Inference algorithms; Information filtering; Information science; Machine learning algorithms; Support vector machine classification; Support vector machines; Testing; Text categorization;

fLanguage

English

Publisher

ieee

Conference_Titel

Innovative Computing, Information and Control, 2006. ICICIC '06. First International Conference on

Conference_Location

Beijing

Print_ISBN

0-7695-2616-0

Type

conf

DOI

10.1109/ICICIC.2006.298

Filename

1691927