Abstract
The automated construction of dynamic system models is an important application area for ILP. We describe a method that learns qualitative models from time-varying physiological signals. The goal is to understand the complexity of the learning task when faced with numerical data, what signal processing techniques are required, and how this affects learning. The qualitative representation is based on Kuipers' QSIM. The learning algorithm for model construction is based on Coiera's GENMODEL. We show that QSIM models are efficiently PAC learnable from positive examples only, and that GENMODEL is an ILP algorithm for efficiently constructing a QSIM model. We describe both GENMOEL which performs RLGG on qualitative states to learn a QSIM model, and the front-end processing and segmenting stages that transform a signal into a set of qualitative states. Next we describe results of experiments on data from six cardiac bypass patients. Useful models were obtained, representing both normal and abnormal physiological states. Model variation across time and across different levels of temporal abstraction and fault tolerance is explored. The assumption made by many previous workers that the abstraction of examples from data can be separated from the learning task is not supported by this study. Firstly, the effects of noise in the numerical data manifest themselves in the qualitative examples. Secondly, the models learned are directly dependent on the initial qualitative abstraction chosen.
Article PDF
Similar content being viewed by others
References
Babaud, J., Witkin, A.P., Baudin, M. & Duda, R.O. 1986. Uniqueness of the Gaussian kernel for scale-space filtering. IEEE Transactions on Pattern Analysis and Machine Intelligence, 8(1):26-33.
Balestra, G. & Liberati, D. 1992. Qualitative simulation of urea extraction during dialysis. IEEE Engineering in Medicine and Biology, 11:80-842.
Blumer, A., Ehrenfeucht, A., Haussler, D. & Warmuth, M.K. 1987. Occam's razor. Information Processing Letters, 24(6):377-380.
Bratko, I., Muggleton, S. & Varšek, A. 1991. Learning qualitative models of dynamic systems. In Proceedings of the International Workshop on Inductive Logic Programming, pages 207-224.
Coiera, E. 1989. Generating qualitative models from example behaviours. DCS Report 8901, Department of Computer Science, University of New South Wales, Sydney, Australia.
Coiera, E. 1992. Monitoring diseases with empirical and model generated histories. Artificial Intelligence in Medicine, 2:135-147, 1990. Revised version appeared in E. Keravnou, editor, Medical Artificial Intelligence I-Deep Models for Medical Knowledge Engineering, pages 71-88, Elsevier, Amsterdam, The Netherlands.
Coiera, E. 1992. Qualitative superposition of unmodelled systems. Technical Report HPL-92-166, Hewlett-Packard Laboratories, Bristol, UK.
Coiera, E. 1993. Editorial: Intelligent monitoring and control of dynamic physiological systems. Artificial Intelligence in Medicine, 5:1-8.
Falkenhainer, B.C. & Michalski, R.S. 1986. Integrating quantitative and qualitative discovery: The ABACUS system. Machine Learning, 1:367-401.
Gobel, F.L., Nordstrom, L.A., Nelson, R.R., Jorgensen, C.R. & Wang, Y. 1978. Rate-pressure product as an index of myocardial oxygen consumption during exercise in patients with angina pectoris. Circulation, 57:549-556.
Guyton, A.C. 1981. Textbook of Medical Physiology. Saunders, Philadelphia, PA, sixth edition.
Hau, D. 1994. Learning qualitative models from physiological signals. Master's thesis, Department of Electrical Engineering and Computer Science, MIT.
Ironi, L., Stefanelli, M. & Lazola, G. 1992. Qualitative models in medical diagnosis. In E. Keravnou, editor, Medical Artificial Intelligence I-Deep Models for Medical Knowledge Engineering. Elsevier, Amsterdam, The Netherlands.
Kearns, M.J. & Vazirani, U.V. 1994. An Introduction to Computational Learning Theory. The MIT Press, Cambridge, MA.
Kuipers, B. 1985. Qualitative simulation in medical physiology: A progress report. Technical Report MIT/LCS/TM-280, Laboratory for Computer Science, MIT, Cambridge, MA.
Kuipers, B. 1986. Qualitative simulation. Artificial Intelligence, 29:289-338.
Kuipers, B. 1987. Qualitative simulation as causal explanation. IEEE Transactions on Systems, Man, and Cybernetics, 17(3):432-444.
Marr, D. & Hildreth, E. 1980. Theory of edge detection. Proc. Roy. Soc. London, 207:187-217.
Muggleton, S., editor. 1992. Inductive Logic Programming. Academic Press, London.
Muggleton, S. 1995. Inverse entailment and Progol. New Generation Computing, 13:245-286.
Oh, T.E., editor. 1990. Intensive Care Manual. Butterworth, Sydney, Australia, third edition.
Oppenheim, A.V. & Schafer, R.W. 1989. Discrete-time Signal Processing. Prentice Hall, Englewood Cliffs, NJ.
Plotkin, G.D. 1971. Automatic Methods of Inductive Inference. PhD thesis, University of Edinburgh.
Quinlan, J.R. 1990. Learning logical definitions from relations. Machine Learning, 5:239-266.
Richards, B.L., Kraan, I. & Kuipers, B.J. 1992. Automatic abduction of qualitative models. In Proceedings of the Tenth National Conference on Artificial Intelligence, pages 723-728.
Rivest, R.L. 1987. Learning decision lists. Machine Learning, 2(3):229-246.
Saidman, L.J. & Smith, N.T., editors. 1984. Monitoring in Anesthesia. Butterworth, Boston, MA, second edition.
Say, R.C. Cem & Kuru, S. 1996. Qualitative system identification: deriving structure from behavior. Artificial Intelligence, 83:75-141.
Schlant, R.C. & Alexander, R.W., editors. 1994. Hurst's the heart: arteries and veins. McGraw-Hill, New York, 8th edition.
Schut, C. & Bredeweg, B. 1996. An overview of approaches to qualitative model construction. Knowledge Engineering Review, 11:1-25.
Siebert, W.M. 1986. Circuits, Signals, and Systems. The MIT Press, Cambridge, MA.
Strang, G. 1989. Wavelets and dilation equations: a brief introduction. SIAM Review, 31(4):614-627.
Toal, P. & Hunter. J. 1990. A qualitative model of cardiac haemodynamics. Technical Report AUCS/TR9001, Department of Computing Science, University of Aberdeen, Aberdeen, U.K.
Valiant, L.G. 1984. A theory of the learnable. Communications of the ACM, 27(11):1134-1142.
Varšek, A. 1991. Qualitative model evolution. In Proceedings of the Twelfth International Joint Conference on Artificial Intelligence, pages 1311-1316, Sydney, Australia.
Weinberg, J., Biswas, G. & Uckun, S. 1990. Continuing adventures in qualitative modelling: A qualitative heart model. In Proceedings of the Third International Conference on Industrial and Engineering Applications of AI Expert Systems (IEA/AIE 90), Charlestown, SC. ACM Press.
Wesseling, K.H., deWit, B., Weber, J.A.P. & Smith, N.T. 1983. A simple device for the continuous measurement of cardiac output. Advances in Cardiovascular Physiology, 5:16-52.
Wyngaarden, J.B., Smith, L.H. & Bennett, J.C., editors. 1992. Cecil Textbook of Medicine. Saunders, Philadelphia, PA, 19th edition.
Yuille, A.L. & Poggio, T.A. 1986. Scaling theorems for zero crossings. IEEE Transactions on Pattern Analysis and Machine Intelligence, 8(1):15-25, January 1986.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Hau, D.T., Coiera, E.W. Learning Qualitative Models of Dynamic Systems. Machine Learning 26, 177–211 (1997). https://doi.org/10.1023/A:1007317323969
Issue Date:
DOI: https://doi.org/10.1023/A:1007317323969