Research Of Two Class Confidence Classification Based On One Class Classifier

Jiang Fangchun; Tian Shengfeng

doi:10.2478/cait-2014-0041

.blurhash-client-img { display: none !important; }

Research Of Two Class Confidence Classification Based On One Class Classifier

Cybernetics and Information Technologies

Volume 14 (2014): Issue 5 (December 2014)

By: Jiang Fangchun and Tian Shengfeng

Open Access

|Dec 2014

Abstract

To have simple and efficient confidence machine learning is an important focus in confidence machine researches. Using one class classifier as a tool, the paper applies it twice for two-class classification problems. Setting reject options and a multi-layer ensemble learning method are used in this study. In this method there is no necessity to set up a specific threshold and the confidence computation is omitted. Realizing five experiments, the study proves it as efficient.

References

1. Vovk, V., A. Gammerman, G. Shafer. Algorithmic Learning in A Random World. USA, Springer Science+Business Media, Inc., 2005.
Search in Google Scholar
2. Liu, C. L., M. Nakagawa. Precise Candidate Selection for Large Character Set Recognition by Confidence Evaluation. - IEEE Trans. on Pattern Analysis and Machine Intelligence, Vol. 22, 2000, No 6, 636-642.10.1109/34.862202
Search in Google Scholar
3. Richard, M. D., R. P. Lippmann. Neural Network Classifiers Estimate Bayesian a Posterior Probabilities. - Neural Computation, Vol. 3, l991, No 4, 461-483.10.1162/neco.1991.3.4.46131167331
Search in Google Scholar
4. Grandvalet, Y. Support Vector Machines with a Reject Option. Vancouver B. C. Canada: Advances in Neural Information Processing Systems 21. - In: Proceedings of the 2008 Conference, 2009, 537-544.
Search in Google Scholar
5. Ping, Zhang, Tien D. Bui, Ching Y. Suen. Anovel Cascade Ensemble Classifier System with a High Recognition Performance on Handwritten Digits. - Pattern Recognition, Vol. 40, 2007, 3415-3429.10.1016/j.patcog.2007.03.022
Search in Google Scholar
6. Kyrki, V., J. K. Kamarainen, H. Kalviainen. Simple Gabor Feature Space for Invariant Object Recognition. - Pattern Recognition Letters, Vol. 25, 2004, No 3, 311-318.10.1016/j.patrec.2003.10.008
Search in Google Scholar
7. Tax, D. M. J. One-Class Classification. Delft, Delft University of Technology, 2001.
Search in Google Scholar
8. Chow, C. K. On Optimum Recognition Error and Reject Tradeoff. - IEEE Trans. on Inf. Theory, Vol. 16, 1970, 41-46.10.1109/TIT.1970.1054406
Search in Google Scholar
9. Zhang, Chunxia, Zhang Jiangsh e. Review of Selective Ensemble Learning Algorithm. - Chinese Journal of Computers, Vol. 34, August 2011, No 8.10.3724/SP.J.1016.2011.01399
Search in Google Scholar
10. Xu, L., A. Krzyzak, C. Y. Suen. Methods of Combining Multiple Classifiers and Their Application to Handwritting Recognition. - IEEE Transaction on Systems, Man, and Cybernetics, Vol. 22, 1992, No 3, 418-435.10.1109/21.155943
Search in Google Scholar
11. Ghodselahi, A. A Hybrid Support Vector Machine Ensemble Model for Credit Scoring. - International Journal of Computer Applications, Vol. 17, 2011, No 5, p. 1.10.5120/2220-2829
Search in Google Scholar
12. http://www.csie.ntu.edu.tw/~cjlin/
Search in Google Scholar
13. Feng, Aimin. Structure-Driven Single Classifier Design and Extension Research. - Nanjing, Nanjing University of Aeronautics and Astronautics, 2011, 3, 14-17.
Search in Google Scholar
14. http://archive.ics.uci.edu/ml/
Search in Google Scholar

Articles in this issue

DOI: https://doi.org/10.2478/cait-2014-0041 | Journal eISSN: 1314-4081 | Journal ISSN: 1311-9702

Journal RSS Feed

Language: English

Page range: 28 - 39

Published on: Dec 30, 2014

Published by: Bulgarian Academy of Sciences, Institute of Information and Communication Technologies

In partnership with: Paradigm Publishing Services

Publication frequency: 4 issues per year

Keywords:

Confidence machine,

credibility,

one class classifier,

ensemble learning,

reject option

Related subjects:

Computer sciences,

Information technology

© 2014 Jiang Fangchun, Tian Shengfeng, published by Bulgarian Academy of Sciences, Institute of Information and Communication Technologies
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Volume 14 (2014): Issue 5 (December 2014)