{"references": ["N. Dimitrova, L. Agnihotri and G. Wei, Video Classification Based on\nHMM Using Text and Face, Proceedings of the European Conference on\nSignal Processing, Finland, 2000", "G. Siyang, L. Quingrui, M. Lin, Meta-classifier in Text Classification,\nhttp://www. comp.nus.edu.sg/~zhouyong/papers/cs5228project.pdf", "W.-H. Lin , A. Houptmann, News Video Classification Using SVMbased\nMultimodal Classifier and Combination Strategies, 2003", "W.-H. Lin , R. Jin, A. Houptmann, A Meta-classification of Multimedia\nClassifiers, International Workshop on Knowledge Discovery in\nMultimedia and Complex Data, Taiwan, 2002", "B. Schoelkopf, A. Smola, \"Learning with Kernels, Support Vector\nMachines\", MIT Press, London, 2002.", "C. Nello, J. Swawe-Taylor, \"An introduction to Support Vector\nMachines\", Cambridge University Press, 2000.", "D. Morariu, L. Vintan, \"A Better Correlation of the SVM kernel-s\nParameters\", Proceeding of the 5th RoEduNet International Conference,\nSibiu, June 2006.", "D. Morariu, L. Vintan, V. Tresp, Feature Selection Methods for an\nImproved SVM Classifier, Proceedings of the 14th International\nConference on Computational and Information Science, pp. 83-89,\nPrague, August 2006", "D. Morariu, L. Vintan, V. Tresp, Evolutionary Feature Selection for Text\nDocuments Using the SVM , Submitted to The 3rd International\nConference on Neural Computing and Patter Recognition, October 2006\n[10] D. Morariu, \"Classification and Clustering using Support Vector\nMachine\", 2nd PhD Report, University \u00d4\u00c7\u00d7Lucian Blaga\" of Sibiu,\nSeptember, 2005, http://webspace.ulbsibiu.ro/ daniel.morariu/html/Docs\n/Report2.pdf.\n[11] Reuters Corpus: http://about.reuters.com/researchandstandards/corpus/.\nReleased in November 2000.\n[12] S. Chakrabarti, \"Mining the Web- Discovering Knowledge from\nhypertext data\", Morgan Kaufmann Press, 2003."]}

Text categorization is the problem of classifying text documents into a set of predefined classes. In this paper, we investigated three approaches to build a meta-classifier in order to increase the classification accuracy. The basic idea is to learn a metaclassifier to optimally select the best component classifier for each data point. The experimental results show that combining classifiers can significantly improve the accuracy of classification and that our meta-classification strategy gives better results than each individual classifier. For 7083 Reuters text documents we obtained a classification accuracies up to 92.04%.