A multi-label Chinese text categorization system based on boosting algorithm

Author

Chen, Junli ; Zhou, Xuezhong ; Wu, Zhaohui

Author_Institution

Coll. of Compute Sci., Zhejiang Univ., China

fYear

2004

fDate

14-16 Sept. 2004

Firstpage

1153

Lastpage

1158

Abstract

This paper presents a multi-label Chinese text categorization system based on Chinese character features and boosting algorithm. This system has been successfully evaluated on the TCM-MED dataset provided by China Academy of traditional Chinese medicine (TCM) and the Reuters-21578 benchmark. We suggest that the TCM-MED dataset can be used as a standard corpus for the Chinese text categorization tasks. We have also carried out experiments to compare the performance of the boosting algorithm with two other traditional algorithms on the same datasets. The results indicate that for the design of a multi-label Chinese text categorization system, the boosting algorithm has a high performance and outperforms the other two algorithms.

Keywords

classification; natural languages; text analysis; China Academy; Chinese character features; Reuters-21578 benchmark; TCM-MED dataset; boosting algorithm; multilabel Chinese text categorization system; traditional Chinese medicine; Algorithm design and analysis; Boosting; Dispatching; Document handling; Educational institutions; Information processing; Machine learning; Machine learning algorithms; Nearest neighbor searches; Text categorization;

fLanguage

English

Publisher

ieee

Conference_Titel

Computer and Information Technology, 2004. CIT '04. The Fourth International Conference on

Print_ISBN

0-7695-2216-5

Type

conf

DOI

10.1109/CIT.2004.1357350

Filename

1357350