ON USING SIMULATION AND STOCHASTIC LEARNING FOR PATTERN RECOGNITION WHEN TRAINING DATA IS UNAVAILABLE - The Case of Disease Outbreak

Dragos Calitoiu, B. John Oommen

2010

Abstract

Pattern Recognition (PR) involves two phases, a Training phase and a Testing Phase. The problems associated with training a classifier when the number of training samples is small are well recorded. Typically, the matrices involved are ill-conditioned and the estimates of the probability distributions are very inaccurate, leading to a very poor classification system. In this paper, we report what we believe are the pioneering results for designing a PR system when there are absolutely no training samples. In such a scenario, we show how we can use a model of the underlying phenomenon and combine it with the principle of stochastic learning to design a very good classifier. By way of example, we consider the case of disease outbreak: Learning the Contagion Parameter in a black box model involving healthy, sick and contagious individuals. The parameter of interest involves Ƞ which is the probability with which an infected person will transmit the disease to a healthy person. Using the theory of Stochastic Point Location (SPL), the problem is reduced to a PR or classification problem in which the SPL is first subjected to a training phase, the outcome of which is used for the testing phase.

References

  1. Chin, J. (2000). Control of Communicable Disease Manual. American Public Health Association, Washington.
  2. Diekmann, O. and Heesterbeek, J. A. P. (2000). Mathematical Epidemiology of Infectious Diseases: Model Building, Analysis and Interpretation. John Wiley.
  3. Greenland, S. (343-357, 1998). Basic methods for sensitivity analysis and external adjustment. In K.J. Rothman KJ and S. Greenland - Modern epidemiology, 2nd ed. Philadelphia, PA: Lippincott-Raven Publishers. Lippinicot Williams & Wilkins.
  4. Lakshmivarahan, S. (1981). Learning Algorithms Theory and Applications. Springer-Verlag.
  5. Narendra, K. S. and Thathachar, M. A. L. (1989). Learning Automata. Prentice-Hall.
  6. Oommen, B. J. (SMC-27B:733-739, 1997). Stochastic searching on the line and its applications to parameter learning in nonlinear optimization. In IEEE Transactions on Systems, Man and Cybernetics.
  7. Oommen, B. J., Kim, S. W., Samuel, M., and Granmo, O. C. (SMC-38B:466-476, 2008.). A solution to the stochastic point location problem in meta-level nonstationary environments. In IEEE Transactions on Systems, Man and Cybernetics.
  8. Oommen, B. J. and Raghunath, G. (SMC-28B:947-954, 1998). Automata learning and intelligent tertiary searching for stochastic point location. In IEEE Transactions on Systems, Man and Cybernetics.
  9. Oommen, B. J., Raghunath, G., and Kuipers, B. (SMC36B:820-836, 2006.). Parameter learning from stochastic teachers and stochastic compulsive liars. In IEEE Transactions on Systems, Man and Cybernetics.
  10. Poznyak, A. S. and Najim, K. (1997). Learning Automata and Stochastic Optimization. Springer-Verlag, Berlin.
  11. Robert, C. P. and Casella, G. (2005). Monte Carlo Statistical Methods (Springer Texts in Statistics). Springer.
  12. Thathachar, M. A. L. T. and Sastry, P. S. (2003). Networks of Learning Automata : Techniques for Online Stochastic Optimization. Kluwer Academic, Boston.
Download


Paper Citation


in Harvard Style

Calitoiu D. and John Oommen B. (2010). ON USING SIMULATION AND STOCHASTIC LEARNING FOR PATTERN RECOGNITION WHEN TRAINING DATA IS UNAVAILABLE - The Case of Disease Outbreak . In Proceedings of the 2nd International Conference on Agents and Artificial Intelligence - Volume 1: ICAART, ISBN 978-989-674-021-4, pages 45-52. DOI: 10.5220/0002716800450052


in Bibtex Style

@conference{icaart10,
author={Dragos Calitoiu and B. John Oommen},
title={ON USING SIMULATION AND STOCHASTIC LEARNING FOR PATTERN RECOGNITION WHEN TRAINING DATA IS UNAVAILABLE - The Case of Disease Outbreak},
booktitle={Proceedings of the 2nd International Conference on Agents and Artificial Intelligence - Volume 1: ICAART,},
year={2010},
pages={45-52},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0002716800450052},
isbn={978-989-674-021-4},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 2nd International Conference on Agents and Artificial Intelligence - Volume 1: ICAART,
TI - ON USING SIMULATION AND STOCHASTIC LEARNING FOR PATTERN RECOGNITION WHEN TRAINING DATA IS UNAVAILABLE - The Case of Disease Outbreak
SN - 978-989-674-021-4
AU - Calitoiu D.
AU - John Oommen B.
PY - 2010
SP - 45
EP - 52
DO - 10.5220/0002716800450052