Title: A Chinese question classification using one-vs-one method as a learning tool

Authors: Yanshan Xiao, Bo Liu, Longbing Cao

Addresses: Guangzhou Asian Games Organising Committee (GAGOC), Guangzhou 510623, P.R. China. ' College of Computer Science and Engineering, South China University of Technology, Guangzhou 510640, P.R. China; Faculty of Information Technology, University of Technology, Sydney, P.O. Box 123, NSW 2007, Australia. ' Faculty of Information Technology, University of Technology, Sydney, P.O. Box 123, NSW 2007, Australia

Abstract: Question classification plays an important role in the question answering system and the errors of question classification will probably result in the failure of question answering. Thus, how to enhance the accuracy is an open question. In order to enhance the accuracies of the Chinese question classification, this paper extends one-against-one method based on SVMs to resolve the problems. The results show the good performance of the algorithm for Chinese question classification problems.

Keywords: support vector machines; SVM; question classification; one-versus-one algorithm; China; question answering systems; information retrieval; information extraction; Chinese.

DOI: 10.1504/IJIIDS.2008.021447

International Journal of Intelligent Information and Database Systems, 2008 Vol.2 No.4, pp.446 - 459

Published online: 27 Nov 2008 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article