Abstract
In this paper, a new method, called EM-EA, is put forward for learning Bayesian network structures from incomplete data. This method combines the EM algorithm with an evolutionary algorithm (EA) and transforms the incomplete data to complete data using EM algorithm and then evolve network structures using the evolutionary algorithm with the complete data. In order to learn Bayesian networks with hidden variables, a new mutation operator has been introduced and the function of the crossover has been correspondingly expanded. The results of the experiments show that EM-EA is more accurate and practical than other network structure learning algorithms that deal with the incomplete data.
This research has been supported by Natural Science Foundation of China, National 973 Fundamental Research Program and 985 Program of Tsinghua University.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Binder, J., Koller, D., Russell, S., Kanazawa, K.: Adaptive probabilistic networks with hidden variables. Machine Learning 29 (1997) 213–244
Cooper, G., Herskovits, E.: A Bayesian method for the induction of probabilistic networks from data. Machine Learning 9 (1992) 309–347
DeJong, K.A., Spears, W.M.: An Analysis of the Interacting Roles of Population Size and Crossover in Genetic Algorithms. Proceedings of the First International Conference on Parallel Problem Solving from Nature (1990), Dortmund, Germany
Friedman, N.: Learning Belief Networks in the Presence of Missing Values and Hidden Variables. Fourteenth International Conference on Machine Learning (ICML’97) (1998), Vanderbilt University, Morgan Kaufmann Publishers
Heckerman, D.: A tutorial on learning Bayesian network. Technical Report MSR-TR-95-06, Microsoft research (1995)
Lam, W., Bacchus, F.: Learning Bayesian belief networks: An approach based on the MDL principle. Computational Intelligence 10 (1994) 269–293
Larranaga, P., Poza, M. et al: Structure Learning of Bayesian Networks by Genetic Algorithms: A Performance Analysis of Control Parameters. IEEE Journal on Pattern Analysis and Machine Intelligence 18(9) (1996) 912–926
Lauritzen, S.L.: The EM algorithm for graphical association models with missing data. Computational Statistics and Data Analysis 19 (1995) 191–201
Myers J.W., Laskey K.B., DeJong K.A.: Learning Bayesian networks from incomplete data using evolutionary algorithms. In GECCO’99 (1999)
Syswerda, G.: Uniform Crossover in Genetic Algorithms. Proceedings of the 3rd International Conference on Genetic Algorithms (1989), Morgan Kaufman
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2001 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Tian, F., Lu, Y., Shi, C. (2001). Learning Bayesian Networks with Hidden Variables Using the Combination of EM and Evolutionary Algorithms. In: Cheung, D., Williams, G.J., Li, Q. (eds) Advances in Knowledge Discovery and Data Mining. PAKDD 2001. Lecture Notes in Computer Science(), vol 2035. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-45357-1_60
Download citation
DOI: https://doi.org/10.1007/3-540-45357-1_60
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-41910-5
Online ISBN: 978-3-540-45357-4
eBook Packages: Springer Book Archive