[Computational-biology] learning bayesian networks with decision
lzhtom at hotmail.com
Wed Aug 17 21:47:37 EST 2005
Now I'm trying to reconstruct a bayesian network from learning dataset.
I want to represent the local conditional probablity distribution as
decision(regression) trees, as illustrated in (Friedman 1996, Learning
bayesian networks with local structure), and (Heckerman 1997, A
bayesian approach to learning bayesian networks with local structure).
However,as the mathematical formulae in the above papers are too abstract
for me,I don't know how to realize them using matlab or R. Could anyone
give me a hint about how to write the code? Or are there any R packages
or matlab toolbox that can do this?
By the way,after reading the two papers mentioned above, I found that
the learning procedure is pretty much the same as building regression
trees, except for the scoring functions. As there already exist functions
for building regression trees in R and matlab, maybe we could modify them
so as to learn local conditional probability distribution with a tree
Best wishes to all!
More information about the Comp-bio