Search found 179 matches
- Thu Jan 06, 2011 4:45 am
- Forum: GeNIe
- Topic: Discretization
- Replies: 3
- Views: 3571
Re: Discretization
No, because in the extreme you will get one state for each value in the data set, which is probably not what you what. You'll have to experiment and somehow evaluate what works best.
- Thu Jan 06, 2011 4:42 am
- Forum: GeNIe
- Topic: Feature Selection algorithm?
- Replies: 2
- Views: 2851
Re: Feature Selection algorithm?
Feature selection uses Greedy Thick Thinning (search forum for more info) to first greedily add features and then remove features that maximizes a Bayesian score function.
- Sun Dec 12, 2010 4:23 am
- Forum: GeNIe
- Topic: Learning DBN parameters (transition probablilities) in GeNIe
- Replies: 20
- Views: 15601
Re: Learning DBN parameters (transition probablilities) in G
You can definitely learn these probabilities from data but you'll need time series data.
- Sun Dec 12, 2010 4:20 am
- Forum: GeNIe
- Topic: Learning Bayesian Network From Data
- Replies: 1
- Views: 2498
Re: Learning Bayesian Network From Data
Is it possible there are no dependencies in the data (i.e., random)?
- Fri Nov 19, 2010 12:49 am
- Forum: SMILE
- Topic: noisy-max question
- Replies: 5
- Views: 5823
Re: noisy-max question
What kind of noisyMax-specific code are you using for parameter learning? When I did some learning on a noisyMax network (in GeNIe) I got the impression that no special code was used. The reason is that after learning, when I saved the network, the noisyMax parameters in the file where replaced by ...
- Thu Oct 21, 2010 4:22 pm
- Forum: GeNIe
- Topic: how to obtain mathematical equation of a network
- Replies: 1
- Views: 2394
Re: how to obtain mathematical equation of a network
Hint: The joint distribution is given by P(C,xi,yi,ti,xf,yf,tf)=P(yi)P(xi|yi)P(ti|yi)P(C|ti)P(tf|C)P(xf|C)P(yf|tf). All other probabilities can be derived from that. Does this help?
- Tue Oct 12, 2010 9:39 pm
- Forum: GeNIe
- Topic: Learning Structure
- Replies: 7
- Views: 5855
Re: Learning Structure
PC is the only option for continuous data, otherwise you need to discretize.
- Tue Oct 12, 2010 9:15 am
- Forum: GeNIe
- Topic: Learning Structure
- Replies: 7
- Views: 5855
Re: Learning Structure
* What is the new structure represent? * How can i build BN from my data (with this new structure representation ), when my main purpose is finding dependency between attributes? The new structure denotes a direct dependency between variables when an edge exists, and edges are oriented if the algor...
- Wed Aug 04, 2010 6:06 am
- Forum: SMILE
- Topic: EM algorithm details
- Replies: 1
- Views: 2226
- Sun Mar 28, 2010 11:43 pm
- Forum: GeNIe
- Topic: Continuous variables
- Replies: 23
- Views: 28336
- Sun Mar 28, 2010 8:57 pm
- Forum: GeNIe
- Topic: Continuous variables
- Replies: 23
- Views: 28336
No, I believe this is still quite different. In case of Bayesian networks only one state can be true at a time but you may not know which one, hence the distributions. In case of a fuzzy Bayesian networks multiple states may be true to certain degrees at the same time, which is conceptually different.
- Sat Mar 27, 2010 5:55 pm
- Forum: GeNIe
- Topic: Continuous variables
- Replies: 23
- Views: 28336
- Tue Jan 05, 2010 4:47 pm
- Forum: GeNIe
- Topic: learning new network
- Replies: 3
- Views: 5542
- Fri Jun 19, 2009 8:31 am
- Forum: SMILE
- Topic: greedythickthinning, what is it exactly?
- Replies: 7
- Views: 6520
- Thu Jun 18, 2009 5:14 pm
- Forum: SMILE
- Topic: greedythickthinning, what is it exactly?
- Replies: 7
- Views: 6520