Automated discovery of empirical equations from data

  • Robert Zembowicz
  • Jan M. Żytkow
Communications Learning and Adaptive Systems
Part of the Lecture Notes in Computer Science book series (LNCS, volume 542)


We describe a machine discovery system for automated finding regularities in numerical data. It can detect a broad range of empirical equations useful in different sciences, and can be easily expanded by addition of new variable transformations. Our system treats experimental error and evaluation of equations in a systematic and statistically sound manner in contradistinction to systems such as BACON, ABACUS, which include error-related parameters, but disregard problems of error analysis and propagation, leading to paradoxical results. Our system propagates error to the transformed variables and assigns error to parameters in equations. Furthermore, it uses errors in weighted least squares fitting, in the evaluation of equations, including their acceptance, rejection and ranking, and uses parameter error to eliminate spurious parameters. In the last part of our paper we analyse the evaluation of equation finding systems. We introduce two convergence tests and we analyze the performance of our system on those tests.


Machine discovery of equations automated curve-fitting 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Eadie, W.T., Drijard, D., James, F.E., Roos, M., Sadoulet, B. 1971. Statistical Methods in Experimental Physics, North-Holland Publishing CompanyGoogle Scholar
  2. Falkenhainer, B.C., & Michalski, R.S. 1986. Integrating Quantitative and Qualitative Discovery: The ABACUS System, Machine Learning, 1, 367–401.Google Scholar
  3. Langley, P., Simon, H.A., Bradshaw, G., & Żytkow J.M. 1987. Scientific Discovery; Computational Exploration of the Creative Processes. Boston, MA: MIT Press.Google Scholar
  4. Nordhausen, B., & Langley, P. 1990. An Integrated Approach to Empirical Discovery. in: J. Shrager & P. Langley eds. Computational Models of Scientific Discovery and Theory Formation, Morgan Kaufmann Publishers, San Mateo, CA, 97–128.Google Scholar
  5. Press, W.H., Flannery, B.P., Teukolsky, S.A., Vetterling, W.T. 1989. Numerical Recipes in Pascal, Cambridge University Press, Cambridge.Google Scholar
  6. Schaffer, C. 1990. A Proven Domain-Independent Scientific Function-Finding Algorithm, Proceedings Eighth National Conference on Artificial Intelligence AAAI Press/The MIT PressGoogle Scholar
  7. Żytkow, J.M. 1987. Combining Many Searches in the FAHRENHEIT discovery system, Proceedings of Fourth International Workshop on Machine Learning Morgan Kaufmann Publ. Los Altos, CA, 281–287.Google Scholar
  8. Żytkow, J.M. & Zhu, J. 1991. Application of Empirical Discovery in Knowledge Acquisition, in: Kodratoff Y. ed. Proceedings of EWSL-91, Springer-Verlag, 101–117.Google Scholar
  9. Żytkow, J.M., Zhu, J. & Hussam, A. 1990. Automated Discovery in a Chemistry Laboratory, Proceedings of the AAAI-90, AAAI Press, 889–894.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1991

Authors and Affiliations

  • Robert Zembowicz
    • 1
  • Jan M. Żytkow
    • 1
  1. 1.Computer Science DepartmentWichita State UniversityWichita

Personalised recommendations