Abstract
The paper presents a concept of a new class of algorithms for discovery of Bayesian networks from data. The basic difficulty of many incremental discovery algorithms in this area is the increasing number of potentially equivalent orientations of edges while an improper choice at the given stage may have dramatic impact on the final network structure. As a remedy, usage of so-called partially oriented graphs at intermediate stages is recommended for which the property of partial dependency separation has been proven. Such partially oriented graphs may maintain in a single structure all equivalent consistent Bayesian networks.
Research partially sponsored under EU project CRIT-2,workpackage Nr.3
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
S. Acid, L.M. de Campos: An algorithm for finding minimum d-separating sets in belief networks, forthcoming, 1996.
G.F.Cooper, E.Herskovits: A Bayesian method for the induction of probabilistic networks from data. Machine Learning 9 (1992), 309–347
D.Geiger, T.Verma, J. Pearl: d-Separation: From theorems to algorithms. W: M.Henrion, R.D.Shachter, L.N.Kamal, J.F.Lemmer (eds): Uncertainty in Artificial Intelligence 5, Elsevier Science Publishers B.V. (North-Holland), 1990, 139–148 288
M.A.Klopotek: Partial Dependency Separation—a new concept for expressing dependence/ independence relations in causal networks. Demonstratio Mathematica. Vol XXXII No 1, 1999, pp. 207–226.
M.A.Klopotek: Methods of Identification and Interpretations of Belief Distributions in the Dempster-Shafer Theory (in Polish), IPI PAN Publisher, Warsaw 1998, ISBN 83–900820–8-x.
W. Lam and F. Bacchus: Learning Bayesian Belief Networks: An Approach based on the MDL Principle, Computational Intelligence, vol. 10, pages 269–293, 1994.
J. Pearl: Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference, Morgan Kaufmann, San Mateo CA, 1988
P. Spirtes, C. Glymour, R. Scheines: Causality from probability. In: G. McKee ed.: Evolving knowledge in natural and artificial intelligence, London: Pitman, 1990, 181–199
P. Spirtes, C. Glymour, R. Scheines: Causation, Prediction and Search, Lecture Notes in Statistics 81, Springer-Verlag, 1993.
T. Urban, T. Kaempke: Recovering dependency graph in uncertain data. M. Mohammadian, ed: Computational Intelligence for Modelling, Control and Automation. Evolutionary Computing & Fuzzy Logic for Intelligent Control, Knowledge Acquisition & Information Retrieval. Concurrent Systems Engineering Series Vol. 55. IOS Press Amsterdam, Berlin, Oxford 1999. pp. 476–481.
Y. Xiang, S.K.M. Wong and N. Cercone: A‘Microscopic’ study of minimum entropy search in learning decomposable Markov networks. Machine Learning, Vol. 26, No. 1, 65–92, 1997.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2000 Physica-Verlag Heidelberg
About this paper
Cite this paper
Kłopotek, M.A., Wierzchoń, S.T. (2000). Discovery of Bayesian Networks from Data with Maintenance of Partially Oriented Graphs. In: Intelligent Information Systems. Advances in Soft Computing, vol 4. Physica, Heidelberg. https://doi.org/10.1007/978-3-7908-1846-8_25
Download citation
DOI: https://doi.org/10.1007/978-3-7908-1846-8_25
Publisher Name: Physica, Heidelberg
Print ISBN: 978-3-7908-1309-8
Online ISBN: 978-3-7908-1846-8
eBook Packages: Springer Book Archive