Eighth ACIS International Conference on Software Engineering, Artificial Intelligence, Networking, and Parallel/Distributed Computing (SNPD 2007) Study on Classification Algorithm of Multi-subject Text Haier International Training Center, Qingdao, China July 30-August 01 ISBN: 0-7695-2909-7
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/SNPD.2007.146
One text may belong to multi-class, but it can not be classified by standard SVM and other approaches. In this paper, a multi-subject text classification algorithm based on fuzzy support vector machines is proposed, 1-a-1 method is used to train sub-classifiers. For the sample to be classified, the sub-classifiers are used to obtain membership matrix, and then according to the sum of every line of membership matrix, the subjects that the sample belongs to can be confirmed. The algorithm was tested on Reuters 21578, the experimental results show that the algorithm has higher performance on recall, precision, and F1.
Citation:
Yu-ping Qin, Qing Ai, Xiu-kun Wang, Xiang-na Li, "Study on Classification Algorithm of Multi-subject Text," snpd, vol. 2, pp.435-438, Eighth ACIS International Conference on Software Engineering, Artificial Intelligence, Networking, and Parallel/Distributed Computing (SNPD 2007), 2007 Usage of this product signifies your acceptance of the Terms of Use. | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb