PRIMEROSE: Probabilistic Rule Induction Method Based on Rough Set Theory

  • Shusaku Tsumoto
  • Hiroshi Tanaka
Conference paper
Part of the Workshops in Computing book series (WORKSHOPS COMP.)

Abstract

Automated knowledge acquisition is an important research issue in machine learning. There have been proposed several methods of inductive learning, such as ID3 family and AQ family. These methods are applied to discover meaningful knowledge from large database, and their usefulness is in some aspects ensured. However, in most of the cases, their methods are of deterministic nature, and reliability of the acquired knowledge is not evaluated statistically, which makes these methods ineffective when applied to the domain of essentially probabilistic nature, such as medical one. Extending concepts of rough set theory to probabilistic domain, we introduce a new approach to knowledge acquisition, which induces probabilistic rules based on rough set theory(PRIMEROSE) and develop a program that extracts rules for an expert system from clinical database, using this method. The results show that the derived rules almost correspond to those of medical experts.

Keywords

Pancreatitis Peri Meningitis 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [1]
    Breiman, L., Freidman, J., Olshen, R., and Stone, C., Classification And Regression Trees. Belmont, CA: Wadsworth International Group, 1984.MATHGoogle Scholar
  2. [2]
    Efron, B., Estimating the error rate of a prediction rule: improvement on cross validation. J. Amer. Statist. Assoc. 78, 316–331, 1983.CrossRefMATHMathSciNetGoogle Scholar
  3. [3]
    Efron, B., How biased is the apparent error rate of a prediction rule ? J. Amer. Statist. Assoc. 82, 171–200, 1986.CrossRefMathSciNetGoogle Scholar
  4. [4]
    Katzberg, J.D. and Ziarko, W. Variable Precision Rough Sets with Asymmetric Bounds, Proceedings of RSKD-93, in this issue.Google Scholar
  5. [5]
    Matsumura, Y., et al., Consultation system for diagnoses of headache and facial pain: RHINOS, Medical Informatics, 11, 145–157, 1986.CrossRefGoogle Scholar
  6. [6]
    McLachlan, G.J., Discriminant Analysis and Statistical Pattern Recognition, John Wiley and Sons, New York, 1992.CrossRefGoogle Scholar
  7. [7]
    Michalski, R.S., A Theory and Methodology of Machine Learning. Michalski, R.S., Carbonell, J.G. and Mitchell, T.M., Machine Learning — An Artificial Intelligence Approach, Morgan Kaufmann, Palo Alto, CA, 1983.Google Scholar
  8. [8]
    Michalski, R.S., et al. The Multi-Purpose Incremental Learning System AQ15 and its Testing Application to Three Medical Domains, in: Proceedings of AAAI-86, 1041–1045, AAAI Press, Palo Alto, CA, 1986.Google Scholar
  9. [9]
    Pawlak, Z., Rough Sets, Kluwer Academic Publishers, 1991.CrossRefMATHGoogle Scholar
  10. [10]
    Quinlan, J.R., Induction of decision trees, Machine Learning, 1, 81–106, 1986.Google Scholar
  11. [11]
    Slowinski, K. et al., Rough sets approach to analysis of data from peritoneal lavage in acute pancreatitis, Medical Informatics, 13, 143–159, 1988.CrossRefGoogle Scholar
  12. [12]
    Walker, M.G. and Olshen, R.A., Probability Estimation for Biomedical Classification Problems. Proceedings of SCAMC-92, McGrawHill, New York, 1992.Google Scholar
  13. [13]
    Ziarko, W., The Discovery, Analysis, and Representation of Data Dependencies in Databases, in:Knowledge Discovery in Database, Morgan Kaufmann, Palo Alto, CA, pp.195–209, 1991.Google Scholar
  14. [14]
    Ziarko, W., Variable Precision Rough Set Model, Journal of Computer and System Sciences, 46, 39–59, 1993.CrossRefMATHMathSciNetGoogle Scholar

Copyright information

© British Computer Society 1994

Authors and Affiliations

  • Shusaku Tsumoto
    • 1
  • Hiroshi Tanaka
    • 1
  1. 1.Department of Informational Medicine Medical Research InstituteTokyo Medical and Dental UniversityBunkyo-ku Tokyo 113Japan

Personalised recommendations