jsat.classifiers.bayesian.graphicalmodel

Class DiscreteBayesNetwork

java.lang.Object

jsat.classifiers.bayesian.graphicalmodel.DiscreteBayesNetwork

All Implemented Interfaces:

Serializable, Cloneable, Classifier

Direct Known Subclasses:

K2NetworkLearner

public class DiscreteBayesNetwork
extends Object
implements Classifier

A class for representing a Baysian Network (BN) for discrete variables. A BN use a graph to representing the relations between variables, and these links are called the structure. The structure of a BN must be specified by an expert using the depends(int, int) method. The target class should be specified as the parent of the variables which have a causal relationship to it. These children of the target class should then have their own children specified. Once the structure has been specified, the network can be trained and used for classification.
If the network structure has not been specified, or has no relationships for the target class, the BN will create an edge from the target class to every variable. If no edges were ever specified, this initialization of edges corresponds to a NaiveBayes implementation.

Author:

Edward Raff

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
protected Map<Integer,ConditionalProbabilityTable>	cpts The Conditional probability tables for each variable
protected DirectedGraph<Integer>	dag The directed Graph that represents this BN
static boolean	DEFAULT_USE_PRIORS Whether or not the classifier should take into account the prior probabilities.
protected CategoricalData	predicting The class we are predicting
protected double[]	priors The prior probabilities of each class value

Constructor Summary

Constructors

Constructor and Description
DiscreteBayesNetwork()

Method Summary

Modifier and Type	Method and Description
CategoricalResults	classify(DataPoint data) Performs classification on the given data point.
Classifier	clone()
void	depends(int parent, int child) Adds a dependency relation ship between two variables that will be in the network.
boolean	supportsWeightedData() Indicates whether the model knows how to train using weighted data points.
void	trainC(ClassificationDataSet dataSet) Trains the classifier and constructs a model for classification using the given data set.
void	trainC(ClassificationDataSet dataSet, ExecutorService threadPool) Trains the classifier and constructs a model for classification using the given data set.

Methods inherited from class java.lang.Object

equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

dag

protected DirectedGraph<Integer> dag

The directed Graph that represents this BN

cpts

protected Map<Integer,ConditionalProbabilityTable> cpts

The Conditional probability tables for each variable

predicting

protected CategoricalData predicting

The class we are predicting

priors

protected double[] priors

The prior probabilities of each class value

DEFAULT_USE_PRIORS

public static final boolean DEFAULT_USE_PRIORS

Whether or not the classifier should take into account the prior probabilities. Default value is true.

See Also:

Constant Field Values

Constructor Detail

DiscreteBayesNetwork

public DiscreteBayesNetwork()

Method Detail

classify

public CategoricalResults classify(DataPoint data)

Description copied from interface: Classifier

Performs classification on the given data point.

Specified by:

classify in interface Classifier

Parameters:

data - the data point to classify

Returns:

the results of the classification.

depends

public void depends(int parent,
                    int child)

Adds a dependency relation ship between two variables that will be in the network. The integer value corresponds the the index of the i'th categorical variable, where the class target's value is the number of categorical variables.

Parameters:

parent - the parent variable, which will be explained in part by the child

child - the child variable, which contributes to the conditional probability of the parent.

trainC

public void trainC(ClassificationDataSet dataSet,
                   ExecutorService threadPool)

Description copied from interface: Classifier

Trains the classifier and constructs a model for classification using the given data set. If the training method knows how, it will used the threadPool to conduct training in parallel. This method will block until the training has completed.

Specified by:

trainC in interface Classifier

Parameters:

dataSet - the data set to train on

threadPool - the source of threads to use.

trainC

public void trainC(ClassificationDataSet dataSet)

Description copied from interface: Classifier

Trains the classifier and constructs a model for classification using the given data set.

Specified by:

trainC in interface Classifier

Parameters:

dataSet - the data set to train on

supportsWeightedData

public boolean supportsWeightedData()

Description copied from interface: Classifier

Indicates whether the model knows how to train using weighted data points. If it does, the model will train assuming the weights. The values returned by this method may change depending on the parameters set for the model.

Specified by:

supportsWeightedData in interface Classifier

Returns:

true if the model supports weighted data, false otherwise

clone

public Classifier clone()

Specified by:

clone in interface Classifier

Overrides:

clone in class Object