Abstract
The attributes that are used in a rule learning task may depend on each other in several ways. Three examples are: two attributes may be associated, the availability of the value of an attribute may depend on another attribute or an attribute may affect the relation between another attribute and the class that is learned. If such dependencies exist, they need a rule language that can express rules dealing with missing values. Also, due to these dependencies there are classes of examples that will not appear as classified training examples and the learning algorithm should be able to handle that. If these dependencies are explicitly known to the system, they can be used to construct an intermediate description language that simplifies and reduces the learning task.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
References
Bundy, A., Silver, B. and Plummer, D. (1985) An anlytical comparison of some rule-learning programs, Artificial Intelligence, 27, p.137–181.
Bundy, A. (1982) The indispensibility of inference in focussing, Internal note Department of Artificial Intelligence, University of Edinburgh.
DeJong, G. and Mooney, A. (1986) Explanation Based Learning: an alternative view, Machine Learning, vol.1, p.145–176.
Flann, N.F. and Dietterich, T. G. (1986) Selecting appropriate representations for learning from examples, Procs. AAAI-86, p. 460–466.
Holte, R.C. (1986) Alternative information structures in incremental learning systems, in: Y. Kodratoff (ed): Procs. EWSL.
Michalski, R.S. (1983) A theory and methodology of inductive learning, in: Michalski e.a. (1983).
Michalski, R.S., Carbonell, J. and Mitchell, T.M. (eds) (1983) Machine learning: an AI approach, Tioga Press.
Mitchell, T.M., (1982), Generalization as search, Artificial Intelligence, 18, p.203–226.
Mitchell, T.M., Keller, R. and Kedar-Cabelli, S.T. (1986) Explanantion based generalization, Machine learning, vol. 1, p.47–80.
Quinlan, J.R. (1987) Decision trees as probabilistic classifiers, in: Procs. 4th IWML, p.31–37.
Rendell, L. (1985) Substantial constructive induction using layered information compression: tractable feature formation in search, in: Proceedings IJCAI-85, Los Angeles.
Rendell, L. (1986) A general framework for induction and a study of selective induction, Machine Learning, vol.1, p.177–226.
van Someren, M.W. (1986) Reducing the description space for rule learning, Procs. ECAI 1986.
Winston, P.H. (1975) Learning structural descriptions from examples, in: Winston, P.H. (ed), The psychology of computer vision, McGraw-Hill.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1989 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
van Someren, M.W. (1989). Using attribute dependencies for rule learning. In: Morik, K. (eds) Knowledge Representation and Organization in Machine Learning. Lecture Notes in Computer Science, vol 347. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0017223
Download citation
DOI: https://doi.org/10.1007/BFb0017223
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-50768-0
Online ISBN: 978-3-540-46081-7
eBook Packages: Springer Book Archive