netkit.classifiers.nonrelational
Class LocalMetaClassifier

java.lang.Object
  netkit.classifiers.ClassifierImp
      netkit.classifiers.nonrelational.LocalMetaClassifier

All Implemented Interfaces:

Classifier, Configurable

Direct Known Subclasses:

MetaMultiplicative

public abstract class LocalMetaClassifier
extends ClassifierImp

Abstract class for combining multiple classifiers. It configures itself with the list of classifiers to use and induces them individually. How to combine them is left to subclasses. It uses the Networklearning.LC_PREFIX property from the configuration details specified in the lclassifier.properties file (in addition to any properties used by the superclass).

Author:

Sofus A. Macskassy (sofmac@gmail.com)

Field Summary

protected java.util.ArrayList<Classifier>

lclassifiers The list of classifiers to combine

Fields inherited from class netkit.classifiers.ClassifierImp

attribute, classPrior, clsIdx, graph, keyIndex, logger, nodeType, right, tmpVector, useIntrinsic, vectorClsIdx

Constructor Summary

LocalMetaClassifier()

Method Summary

void	configure(Configuration config) Configures the classifier by getting the list of classifiers to use (comma-separated list in the NetworkLearning.LC_PREFIX property.
protected java.lang.String	getClassifierNames()
Configuration	getDefaultConfiguration() Default configuration uses only the naive Bayes classifier in addition to any defaults from the superclass
void	induceModel(Graph graph, DataSplit split) This induces each of the local classifiers individually.

Methods inherited from class netkit.classifiers.ClassifierImp

addListener, classify, classify, clearListeners, estimate, estimate, getAttributeNames, getLogger, getNofifyListeners, makeVector, notifyListeners, notifyListeners, removeListener, reset, setNofityListeners

Methods inherited from class java.lang.Object

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

Methods inherited from interface netkit.classifiers.Classifier

estimate, getDescription, getName, getShortName

Field Detail

lclassifiers

protected java.util.ArrayList<Classifier> lclassifiers

The list of classifiers to combine

Constructor Detail

LocalMetaClassifier

public LocalMetaClassifier()

Method Detail

getDefaultConfiguration

public Configuration getDefaultConfiguration()

Default configuration uses only the naive Bayes classifier in addition to any defaults from the superclass

Specified by:

getDefaultConfiguration in interface Configurable

Overrides:

getDefaultConfiguration in class ClassifierImp

Returns:

a default Configuration object

configure

public void configure(Configuration config)

Configures the classifier by getting the list of classifiers to use (comma-separated list in the NetworkLearning.LC_PREFIX property.

Specified by:

configure in interface Configurable

Overrides:

configure in class ClassifierImp

Parameters:

config - The configuration object to use to configure this classifier.

See Also:

netkit.classifiers.NetworkLearning.LC_PREFIX

induceModel

public void induceModel(Graph graph,
                        DataSplit split)

This induces each of the local classifiers individually. If learning need be done to learn how to combine them, then that needs to be done by subclasses.

Specified by:

induceModel in interface Classifier

Overrides:

induceModel in class ClassifierImp

Parameters:

graph - Graph whose nodes are to be estimated

split - The split between training and test. Used to get the nodetype and class attribute.

getClassifierNames

protected java.lang.String getClassifierNames()

Returns:

the list of classifier names from the configuration object when configure was called

See Also:

configure(netkit.util.Configuration)