Search found 1474 matches

by shooltz[BayesFusion]
Thu May 22, 2008 7:46 pm
Forum: SMILE
Topic: diagnosis network functions in GeNie / SMILE
Replies: 6
Views: 13903

Re: diagnosis network functions in GeNie / SMILE

Can you post the actual network (not just the screenshot) here?
by shooltz[BayesFusion]
Wed May 21, 2008 5:53 pm
Forum: GeNIe
Topic: problems building networks based on a database
Replies: 18
Views: 40565

There are two parameters for controlling EM algorithm:

1) 'confidence level'. It's described in the dialog box shown after matching network and data and before running EM; see this thread for more information:

http://genie.sis.pitt.edu/forum/viewtopic.php?t=4

2) 'randomize initial parameters ...
by shooltz[BayesFusion]
Wed May 21, 2008 5:08 pm
Forum: GeNIe
Topic: problems building networks based on a database
Replies: 18
Views: 40565

What about using one of these sites where you type the search term and they return relevant links?
by shooltz[BayesFusion]
Wed May 21, 2008 3:53 pm
Forum: GeNIe
Topic: problems building networks based on a database
Replies: 18
Views: 40565

Check this and/or GeNIe's help file:

http://genie.sis.pitt.edu/wiki/GeNIe_Tu ... Parameters
by shooltz[BayesFusion]
Wed May 21, 2008 11:42 am
Forum: GeNIe
Topic: problems building networks based on a database
Replies: 18
Views: 40565

The naive Bayes algorithm can't be applied in your case - it creates simple model with arcs directed from 'class node' into all other nodes. Since you're interested in structure where arcs go into specified node (Off_Alw_Fo), the PC algorithm should work. In fact, you've obtained the model which is ...
by shooltz[BayesFusion]
Wed May 21, 2008 10:54 am
Forum: SMILE
Topic: [.NET]libary for .NET compact framework
Replies: 1
Views: 5651

Re: [.NET]libary for .NET compact framework

Would you please to provide library .NET compact framework? Already having a library for .NET framework, is it hard job to convert .NET framework to .NET compact framework?

I don't know - we don't have any experience with .NET CF. You're the first user asking for support on this platform. While ...
by shooltz[BayesFusion]
Mon May 19, 2008 9:03 am
Forum: GeNIe
Topic: problems building networks based on a database
Replies: 18
Views: 40565

jam wrote:The parameter naive doing everything contrary to what we wish to undertake, and i do not understand why.
Can you post your data/learned network and explain what's the problem with the net?
by shooltz[BayesFusion]
Fri May 16, 2008 10:55 am
Forum: SMILE
Topic: Bugs with decision nodes as parents of nature nodes
Replies: 3
Views: 9997

Re: Bugs with decision nodes as parents of nature nodes

It appears that there is a bug with decision nodes as parent of nature nodes in the Smile engine. Please import the following Netica file to see the bug.

The file loads OK with the following warning messages logged (despite the 'ERROR' prefix, they're just warnings):

(line 13) : ERROR ...
by shooltz[BayesFusion]
Fri May 16, 2008 9:56 am
Forum: GeNIe
Topic: problems building networks based on a database
Replies: 18
Views: 40565

On the oder hand, , in the help of genie appears one thing, in my software appears otherwise, I would like to know why. Additions a document

The screenshot in help file shows old dialog, which had all parameters for all structure learning algorithms visible. They were just enabled/disabled ...
by shooltz[BayesFusion]
Fri May 16, 2008 9:38 am
Forum: SMILE
Topic: network width/height
Replies: 1
Views: 6714

Re: network width/height

Is there a way to find network boundaries (i.e. maxX and maxY)?

Iterate over the nodes, retrieve the position of each node and create the union rectangle.

If you're using C++, then you could use DSL_network::SimpleGraphLayout after adding the nodes at any position. This would use parent ordering ...
by shooltz[BayesFusion]
Thu May 15, 2008 7:03 pm
Forum: GeNIe
Topic: problems building networks based on a database
Replies: 18
Views: 40565

jam wrote:I want to know that other options exist apart from the background knowledge and parameter naive to build my network without the Genie randomization
What do you mean by 'parameter naive'?
by shooltz[BayesFusion]
Thu May 15, 2008 6:12 pm
Forum: GeNIe
Topic: problems building networks based on a database
Replies: 18
Views: 40565

Re: problems building networks based on a database

I'm not sure what exactly is your problem. Did you used automatic translator? If so, please ask somebody who knows English to translate for you.
by shooltz[BayesFusion]
Thu May 15, 2008 1:25 pm
Forum: GeNIe
Topic: problems building networks based on a database
Replies: 18
Views: 40565

Re: problems building networks based on a database

I have tried to build a Bayesian network from a database, apparently genie takes the data and unites them at random

GeNIe certainly does not 'unite data at random'. All available data is read and used by learning algorithms.

The naive algorithm doesn't allow any background knowledge, but you can ...
by shooltz[BayesFusion]
Thu May 15, 2008 8:57 am
Forum: SMILE
Topic: Bugs with decision nodes as parents of nature nodes
Replies: 3
Views: 9997

Re: Bugs with decision nodes as parents of nature nodes

andrew_ling wrote:Please import the following Netica file to see the bug.
There's no file attached to your post.
by shooltz[BayesFusion]
Fri May 09, 2008 11:16 am
Forum: SMILE
Topic: Error -5 running EPIS
Replies: 1
Views: 6250

Re: Error -5 running EPIS

Can you perhaps elaborate on the circumstances under which this error is likely to occur?

(disclaimer: not a sampling expert here)

Most likely reason is that probability of the evidence you've set is very small. Did you increase the number of samples above the default value with Network ...