Advances in Atmospheric Sciences

, Volume 13, Issue 4, pp 471–488 | Cite as

Machine Learning of Weather Forecasting Rules from Large Meteorological Data Bases

  • Honghua Dai


Discovery of useful forecasting rules from observational weather data is an outstanding interesting topic. The traditional methods of acquiring forecasting knowledge are manual analysis and investigation performed by human scientists. This paper presents the experimental results of an automatic machine learning system which derives forecasting rules from real observational data. We tested the system on the two large real data sets from the areas of central China and Victoria of Australia. The experimental results show that the forecasting rules discovered by the system are very competitive to human experts. The forecasting accuracy rates are 86.4% and 78% of the two data sets respectively.

Key words

Weather forecasting Machine learning Machine discovery Meteorological expert system Meteorological knowledge processing Automatic forecasting 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Gordon Allen, Victor Ciesielski, and Warwick Bolam (1986), Evaluation of an expert system for predicting rain in Melbourne, In Proceedings of First Australian Artificial Intelligence Congress, pages 16–117, Melbourne.Google Scholar
  2. Gordon Allen and J.F. Le Marshall (1993), An evaluation of neural networks and discriminant analysis methods for application in operational rain forecasting. Technical report, Bureau of Meteorology Research Center, Melbourne, Australia.Google Scholar
  3. Booker L.B., D.E. Goldberg, and J.H. Holland (1989), Classifier systems and genetic algorithms, Artificial Intelligence, 46(1–3): 235–282.CrossRefGoogle Scholar
  4. Carbonell J.G, (1989), Introduction: Paradigms for machine learning, Artificial Intelligence, 40(1–3):1–9.CrossRefGoogle Scholar
  5. Victor Ciesielski and Honghua Dai (1994), FISHERMAN: a comprehensive discovery, learning and forecasting systems, In Proceedings of 2nd Singapore International Conference on Intelligent System, pages B297 (l)-B297(6).Google Scholar
  6. Cheeseman P., J. Kelly, M. Self, J. Stutz, W. Taylor, and D. Treeman (1988), AUTOCLASS: A Bayesian Classification System, Technical report, NASA Ames Research Centre.Google Scholar
  7. Dai Honghua (1985), The development of the WMES system and a study to its theoretical problems, M. Sc. Thesis, IAP of Academia Sinica, 1(1). Google Scholar
  8. Dai Honghua (1986), The Foundation to Meteorological Expert Systems, IAP, Academia Sinica, Beijing.Google Scholar
  9. Dai Honghua (1986), The regional heavy rain forecasting expert system—the WMES system, China Meteorology, 4(2)(in Chinese).Google Scholar
  10. Dai Honghua (1994), Learning of forecasting rules from large noisy real meteorological data, PhD. Dissertation, Department of Computer Science, RMIT. Google Scholar
  11. Dai Honghua and Victor Ciesielski (1992), The application of machine learning and machine discovery to atmospheric science, In Proceedings of The 4th International Summer Colloquium and the International Sysmposium on Climate, Environment of Geophysical Fluid Dynamics, volume 1, pages 446–462, 20–25 July 1992, China Meteorological Press, Beijing.Google Scholar
  12. Dai Honghua and Victor Ciesielski (1994), Learning of inexact rules by the FISH-NET algorithm from low quality data, In Proceedings of 7th Australian Joint Conference on Artificial Intelligence, pages 108–115, World Scientific, Nov, 1994.Google Scholar
  13. Dai Honghua and Victor Ciesielski (1994), The low prediction accuracy problem in learning, In Proceeding of Second Australian and New Zealand Conference On Intelligent Systems, pages 367–371, IEEE, Nov,, 1994.Google Scholar
  14. Dai Honghua and Victor Ciesielski (1995), Inexact field learning using the FISH-NET algorithm, Technical report, Department of Computer Science.Google Scholar
  15. Dai Honghua, Victor Ciesielski, and Zhuopeng Feng (1993), Successive constrained discovery of forecasting rules from very large data bases, In Proceedings of third International Conference for Computer Scientists, ICYCS’93, pages 10.12-10.15, Tsinghua University Press, July 15–17, 1993, Beijing.Google Scholar
  16. G.F. DeJong and R. Mooney (1986), Explanation—based learning: An alternative view, Machine Learning, 1(1986) (1): 145–176.Google Scholar
  17. Dai Honghua, Qi Song Zheng, and Zhao Xi Zhao (1987), An expert system for predicting heavy rain, Advances in Atmospheric Sciences, 4(4): 496–505.CrossRefGoogle Scholar
  18. Hinton G.E. (1989), Connectionist learning procedures. Artificial Intelligence, 40(1-3): 185–234.CrossRefGoogle Scholar
  19. Keith K.Y. Lam (1993), Hybridized k—nearest neighbours and genetic algorithms for rainfall prediciton, Technical report, Department of Computer Science, Royal Melbourne Institute of Technology.Google Scholar
  20. Craig A. Lindley, Charles Chung, and Ramkuman Venkatachalam (1993), An evaluation of automatic and manual knowledge acquisition techniques for weather forecast operations, Technical report, CSIRO Division of Information Technology, Australia.Google Scholar
  21. Moninger, W.R., J. Bullas, B. deLorenzis, E. Ellison, J. Flueck, J.C. McLeod, C. Lusk, P.D. Lampru, R.S. Philips, W.F. Roberts, R. haw, T.R. Stewart, S. Weaver, K.C. Young, and S.M. Zubrick (1991), Shoot-89, a comparative evaluation of knowledge-based systems that forecast severe weather, Bulletin American Meteorological Society, pages 1339–1354.Google Scholar
  22. Mitchell Tom M. (1986), Explanation-based generalization: A unifying view, Machine Learning, 1(1): 47–80.Google Scholar
  23. Moninger W.R., (1988), ARCHER: A prototype expert system for identifying some meteorological phenomena, Journal of Atmospheric and Oceanic Technology, 50(1): 144–148.CrossRefGoogle Scholar
  24. Gregory Piatetsky-Shapiro and William J. Frawiey (1991), Knowledge Discovery in Databases, The MIT Press.Google Scholar
  25. Ross Quinlan (1986), Induction of decision trees. Machine Learning, 1: 81–106.Google Scholar
  26. Ross Quinlan (1992), C4.5: Programms fer Machine Learning, Morgan Kaufmann Publishers, San Mateo, California, USA.Google Scholar
  27. Roberts W.F., W.R. Moninger, B. deLorenzis, E. Ellision, J. Flueck, J.C. McLeod, C. Lusk, P.D. Lampru, R. Shaw, and T.R. Steward (1989), A field test of ai systems applied to the problem of severe weather forecasting: SHOOTOUT-89, Technical report, American Meteorological Society, Boston.Google Scholar
  28. Roberts W.F. (1988), Severe weather advisory program: A dynamically adjusted rule-based system for severe weather forecasting, 4th international conference on interactive information and processing system for meteorology, Oceanography and Hydrology. Google Scholar
  29. Jude W. Shavlik (1990), Readings in Machine Learning, Morgan Kaufmann Publishers, INC., San Mateo, California.Google Scholar
  30. Paul E. Utgoff (1938), 1D5: An incremental 1D3, Proceedings of the Fifih International Conference On Machine Learning, June 12–14, 1988.Google Scholar
  31. Sholom M. Weiss and Casimir A. Kulikowski (1991), Computer Systems that Learn: Classification and Prediction Methods fi-om Statistics, Neural Nets, Machine Learning and Expert Systems, Morgan Kaufmann Publishers, Palo Alto, 1991.Google Scholar

Copyright information

© the editorial office of Advancees in Atmospheric Sciences (China) 1996

Authors and Affiliations

  • Honghua Dai
    • 1
  1. 1.Department of Computer ScienceMonasb UniversityAustralia

Personalised recommendations