--On Wednesday, November 24, 1999, 10:48 AM -0800 James Cussens
<[EMAIL PROTECTED]> wrote:
> I've just finished giving a few lectures on join tree propagation, and
> it struck me that some sort of animated demonstration of how it works
> would be a lot more illuminating than the static descriptions I have
> been using. I was thinking of something that eg showed the messages
> being constructed and absorbed. Most packages that I have looked at
> quite naturally hide this stuff away - but for teaching you want it
> made explicit.
>
> Before I recruit some unsuspecting student to produce said software as
> a summer project, I though I had better check that there is not
> something suitable already.
>
> (Btw, I'm way too lazy to produce my own animations with PowerPoint or
> StarOffice.)
Jim,
I know that it is by far not what you want, but one thing that I use in
teaching is the capability of GeNIe to show on the screen whether the
posterior probability distribution of a node is invalid (a small question
mark icon in the lower right corner of the node). This is an element of
GeNIe's relevance reasoning/lazy evaluation scheme. When you update the
model, all question marks disappear. Observing evidence invalidates the
posterior probability distribution over all nodes that are d-connected to
it and this is shown by question marks. This shows (in an admittedly crude
way) how evidence spreads through the network. You can switch this off
(Network/Update immediately), but I often use this mode in teaching. By
the way, this all is not specific to the join tree propagation. I suggest
that you also look at the capability of Hugin and Netica to display the
junction tree -- it may be useful in explaining the clustering algorithm.
Cheers,
Marek
--------------------------------------------------------------------------
Marek J. Druzdzel http://www.pitt.edu/~druzdzel
