Re: BNs from Decision Trees

Ian Miller (imiller@csd.abdn.ac.uk)
Wed, 23 Sep 1998 10:05:22 +0100 (BST)

Lars M. Nielsen wrote:
>
> Michael Mayo wrote:
> >
> > If I have a decision tree with probabilities attached to each and
> > every outcome arc and leaf nodes representing conclusions, are there
> > any algorithms for inducing a bayesian network?
>
[...]
>
> If there are only a few irregularities in the decision tree, you can do
> some tricks to get around it when converting to a Bayesian network
> (well, you also can if there are many, but the Bayesian network would be
> ugly).
>
> In these in-between cases, it might be interesting to be able to combine
> the two modelling techniques. I don't know if anybody has tried to do
> that!?

Ron Kohavi's NBtree algorithm induces decision trees with naive Bayes
classifiers at the leaves:-

http://robotics.stanford.edu/~ronnyk/nbtree.ps.gz

-- 
+-------------------------------------+---------------------------------------+
| ian miller; research student;       | When the Master governs, the people   |
| learning agents and systems group;  | are hardly aware that he exists.      |
| university of aberdeen, scotland.   |    ... When his work is done,         |
+-------------------------------------+        the people say, "Amazing:      |
| imiller@csd.abdn.ac.uk              |        we did it, all by ourselves!"  |
| http://www.csd.abdn.ac.uk/~imiller/ |                             - Lao Tzu |
+-------------------------------------+---------------------------------------+