ARTIFICIAL NEURAL NETWORKS LEARNING IN ROC SPACE

Cristiano Leite Castro, Antônio Padua Braga

2009

Abstract

In order to control the trade-off between sensitivity and specificity of MLP binary classifiers, we extended the Backpropagation algorithm, in batch mode, to incorporate different misclassification costs via separation of the global mean squared error between positive and negative classes. By achieving different solutions in ROC space, our algorithm improved the MLP classifier performance on imbalanced training sets. In our experiments, standard MLP and SVM algorithms were compared to our solution using real world imbalanced applications. The results demonstrated the efficiency of our approach to increase the number of correct positive classifications and improve the balance between sensitivity and specificity.

References

  1. Asuncion, A. and Newman, D. (2007). UCI machine learning repository.
  2. Cortes, C. and Mohri, M. (2004). Auc optimization vs. error rate minimization. In Advances in Neural Information Processing Systems 16. MIT Press, Cambridge, MA.
  3. Cortes, C. and Vapnik, V. (1995). Support-vector networks. Mach. Learn., 20(3):273-297.
  4. Egan, J. P. (1975). Signal Detection Theory and ROC Analysis. Academic Press.
  5. Elkan, C. (2001). The foundations of cost-sensitive learning. In Proceedings of the Seventeenth International Joint Conference on Artificial Intelligence, IJCAI, pages 973-978.
  6. Everson, R. M. and Fieldsend, J. E. (2006). Multi-class roc analysis from a multi-objective optimisation perspective. Pattern Recogn. Lett., 27(8):918-927.
  7. Fawcett, T. (2004). Roc graphs: Notes and practical considerations for researchers. Technical report.
  8. Graening, L., Jin, Y., and Sendhoff, B. (2006). Generalization improvement in multi-objective learning. In International Joint Conference on Neural Networks, pages 9893-9900. IEEE Press.
  9. Haykin, S. (1994). Neural networks: A comprehensive foundation. Macmillan, New York.
  10. Japkowicz, N. (2000). The class imbalance problem: Significance and strategies. In Proceedings of the 2000 International Conference on Artificial Intelligence, pages 111-117.
  11. Kupinski, M. A. and Anastasio, M. A. (1999). Multiobjective genetic optimization of diagnostic classifiers with implications for generating receiver operating characterisitic curves. IEEE Trans. Med. Imag., 18:675-685.
  12. Provost, F. and Fawcett, T. (2001). Robust classification for imprecise environments. Mach. Learn., 42(3):203- 231.
  13. Provost, F. J., Fawcett, T., and Kohavi, R. (1998). The case against accuracy estimation for comparing induction algorithms. In ICML 7898: Proceedings of the Fifteenth International Conference on Machine Learning, pages 445-453, San Francisco, CA, USA. Morgan Kaufmann Publishers Inc.
  14. Rumelhart, D. E. and McClelland, J. L. (1986). Parallel distributed processing: Explorations in the microstructure of cognition, volume 1: Foundations. MIT Press.
  15. Sanchez, M. S., Ortiz, M. C., Sarabia, L. A., and Lleti, R. (2005). On pareto-optimal fronts for deciding about sensitivity and specificity in class-modelling problems. Analytica Chimica Acta, 544(1-2):236-245.
  16. Sun, Y., Kamel, M. S., Wong, A. K. C., and Wang, Y. (2007). Cost-sensitive boosting for classification of imbalanced data. Pattern Recognition, 40(12):3358- 3378.
  17. Van Gestel, T., Suykens, J. A. K., Baesens, B., Viaene, S., Vanthienen, J., Dedene, G., De Moor, B., and Vandewalle, J. (2004). Benchmarking least squares support vector machine classifiers. Mach. Learn., 54(1):5-32.
  18. Weiss, G. M. (2004). Mining with rarity: a unifying framework. SIGKDD Explor. Newsl., 6(1):7-19.
  19. Wu, G. and Chang, E. Y. (2005). Kba: Kernel boundary alignment considering imbalanced data distribution. IEEE Trans. Knowl. Data Eng., 17(6):786-795.
  20. Zhou, Z.-H. and Liu, X.-Y. (2006). Training cost-sensitive neural networks with methods addressing the class imbalance problem. IEEE Trans. Knowl. Data Eng., 18(1):63-77.
Download


Paper Citation


in Harvard Style

Leite Castro C. and Padua Braga A. (2009). ARTIFICIAL NEURAL NETWORKS LEARNING IN ROC SPACE . In Proceedings of the International Joint Conference on Computational Intelligence - Volume 1: ICNC, (IJCCI 2009) ISBN 978-989-674-014-6, pages 484-489. DOI: 10.5220/0002324404840489


in Bibtex Style

@conference{icnc09,
author={Cristiano Leite Castro and Antônio Padua Braga},
title={ARTIFICIAL NEURAL NETWORKS LEARNING IN ROC SPACE},
booktitle={Proceedings of the International Joint Conference on Computational Intelligence - Volume 1: ICNC, (IJCCI 2009)},
year={2009},
pages={484-489},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0002324404840489},
isbn={978-989-674-014-6},
}


in EndNote Style

TY - CONF
JO - Proceedings of the International Joint Conference on Computational Intelligence - Volume 1: ICNC, (IJCCI 2009)
TI - ARTIFICIAL NEURAL NETWORKS LEARNING IN ROC SPACE
SN - 978-989-674-014-6
AU - Leite Castro C.
AU - Padua Braga A.
PY - 2009
SP - 484
EP - 489
DO - 10.5220/0002324404840489