|
||||||||||
PREV CLASS NEXT CLASS | FRAMES NO FRAMES | |||||||||
SUMMARY: NESTED | FIELD | CONSTR | METHOD | DETAIL: FIELD | CONSTR | METHOD |
java.lang.Objectweka.classifiers.Classifier
weka.classifiers.DistributionClassifier
weka.classifiers.bayes.BayesNet
weka.classifiers.bayes.BayesNetB
weka.classifiers.bayes.BayesNetB2
Class for a Bayes Network classifier based on Buntines hill climbing algorithm for learning structure, but augmented to allow arc reversal as an operation. Works with nominal variables only.
Field Summary |
Fields inherited from class weka.classifiers.bayes.BayesNet |
m_Distributions, m_Instances, m_nOrder, m_NumClasses, m_ParentSets, TAGS_SCORE_TYPE |
Constructor Summary | |
BayesNetB2()
|
Method Summary | |
void |
buildStructure()
buildStructure determines the network structure/graph of the network with Buntines greedy hill climbing algorithm, restricted by its initial structure (which can be an empty graph, or a Naive Bayes graph. |
static void |
main(java.lang.String[] argv)
Main method for testing this class. |
Methods inherited from class weka.classifiers.bayes.BayesNetB |
globalInfo |
Methods inherited from class weka.classifiers.DistributionClassifier |
calculateEntropy, calculateLabeledInstanceMargin, calculateMargin, classifyInstance |
Methods inherited from class weka.classifiers.Classifier |
forName, makeCopies |
Methods inherited from class java.lang.Object |
clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait |
Constructor Detail |
public BayesNetB2()
Method Detail |
public void buildStructure() throws java.lang.Exception
buildStructure
in class BayesNetB
java.lang.Exception
public static void main(java.lang.String[] argv)
argv
- the options
|
||||||||||
PREV CLASS NEXT CLASS | FRAMES NO FRAMES | |||||||||
SUMMARY: NESTED | FIELD | CONSTR | METHOD | DETAIL: FIELD | CONSTR | METHOD |