|
This Bayes Network learning algorithm uses a hill climbing algorithm restricted by an order on the variables.
For more information see:
G.F. Cooper, E. Herskovits (1990). A Bayesian method for constructing Bayesian belief networks from databases.
G. Cooper, E. Herskovits (1992). A Bayesian method for the induction of probabilistic networks from data. Machine Learning. 9(4):309-347.
Works with nominal variables and no missing values only.
BibTeX:
@proceedings{Cooper1990,
author = {G.F. Cooper and E. Herskovits},
booktitle = {Proceedings of the Conference on Uncertainty in AI},
pages = {86-94},
title = {A Bayesian method for constructing Bayesian belief networks from databases},
year = {1990}
}
@article{Cooper1992,
author = {G. Cooper and E. Herskovits},
journal = {Machine Learning},
number = {4},
pages = {309-347},
title = {A Bayesian method for the induction of probabilistic networks from data},
volume = {9},
year = {1992}
}
Valid options are:
-N
Initial structure is empty (instead of Naive Bayes)
-P <nr of parents>
Maximum number of parents
-R
Random order.
(default false)
-mbc
Applies a Markov Blanket correction to the network structure,
after a network structure is learned. This ensures that all
nodes in the network are part of the Markov blanket of the
classifier node.
-S [BAYES|MDL|ENTROPY|AIC|CROSS_CLASSIC|CROSS_BAYES]
Score type (BAYES, BDeu, MDL, ENTROPY and AIC)
author: Remco Bouckaert (rrb@xm.co.nz) version: $Revision: 1.6 $ |