[R--gR] deal: applying the network

[Claus Dethlefsen / Aalborg Sygehus]

Dear Sam
 
Deal supports the .net format for Bayesian networks, which can be read by Hugin. There is a (free) light version of Hugin which can be used for 'toy examples'. However, there are many other software packages, some of which also support the .net format, see
 
http://www.cs.ubc.ca/~murphyk/Bayes/bnsoft.html
 
Deal has only the 'learning' capability and inference must be done using methods outside of 'deal'.
 
Best,
Claus
 
________________________________
Claus Dethlefsen, Msc, PhD
Statistiker ved Kardiovaskulært Forskningscenter

 
Forskningens Hus
Aalborg Sygehus 
Sdr. Skovvej 15
9000 Aalborg

Tlf:   9932 6863
email: aas.claus.dethlefsen at nja.dk <mailto:aas.claus.dethlefsen at nja.dk> 

________________________________

Fra: r-sig-gr-bounces at stat.math.ethz.ch på vegne af Sam Steingold
Sendt: ti 21-03-2006 17:34
Til: Claus Dethlefsen / Aalborg Sygehus
Cc: r-sig-gr at stat.math.ethz.ch
Emne: Re: [R--gR] deal: applying the network



Dear Claus

> * Claus Dethlefsen / Aalborg Sygehus <nnf.pynhf.qrguyrsfra at awn.qx> [2006-03-21 09:24:28 +0100]:
>
> To use the trained network together with evidence, you need to
> transfer the network to Hugin.

but hugin is a commercial product.
are you saying that I cannot use the network built by deal without buying
Hugin?!

how do I evaluate a network built by deal?
e.g., the residuals? error rate (for discrete nodes)?

> There is no way to incremently update the network, so you would need
> to relearn the whole thing unfortunately.
ok.

thanks!

--
Sam Steingold (http://www.podval.org/~sds) on Fedora Core release 4 (Stentz)
http://www.honestreporting.com http://www.mideasttruth.com http://ffii.org
http://www.dhimmi.com http://www.savegushkatif.org http://www.memri.org
It's not just a language, it's an adventure.  Common Lisp.

_______________________________________________
R-sig-gR mailing list
R-sig-gR at stat.math.ethz.ch
https://stat.ethz.ch/mailman/listinfo/r-sig-gr