mulan.classifier.lazy
Class MultiLabelKNN

java.lang.Object
  mulan.classifier.MultiLabelLearnerBase
      mulan.classifier.lazy.MultiLabelKNN

All Implemented Interfaces:

Serializable, MultiLabelLearner, TechnicalInformationHandler

Direct Known Subclasses:

BRkNN, IBLR_ML, MLkNN

public abstract class MultiLabelKNN
extends MultiLabelLearnerBase

Superclass of all KNN based multi-label algorithms

Author:

Eleftherios Spyromitros-Xioufis, Grigorios Tsoumakas

See Also:

Serialized Form

Field Summary

protected DistanceFunction	dfunc Implementing Euclidean distance (or similarity) function.
protected int	distanceWeighting Whether the neighbors should be distance-weighted.
protected LinearNNSearch	lnn Class implementing the brute force search algorithm for nearest neighbor search.
protected int	numOfNeighbors Number of neighbors used in the k-nearest neighbor algorithm
protected Instances	train The training instances
static int	WEIGHT_INVERSE weight by 1/distance.
static int	WEIGHT_NONE no weighting.
static int	WEIGHT_SIMILARITY weight by 1-distance.

Fields inherited from class mulan.classifier.MultiLabelLearnerBase

featureIndices, labelIndices, numLabels

Constructor Summary

MultiLabelKNN()
The default constructor

MultiLabelKNN(int numOfNeighbors)
Initializes the number of neighbors

Method Summary

protected void	buildInternal(MultiLabelInstances trainSet) Learner specific implementation of building the model from MultiLabelInstances training data set.
boolean	isUpdatable() Returns value indicating if learner is updatable, so if learner is able to perform on-line learning.
void	setDfunc(DistanceFunction dfunc) Sets a distance function
void	setDistanceWeighting(int distanceWeighting)

Methods inherited from class mulan.classifier.MultiLabelLearnerBase

build, debug, getDebug, getTechnicalInformation, globalInfo, isModelInitialized, makeCopy, makePrediction, makePredictionInternal, setDebug

Methods inherited from class java.lang.Object

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

Field Detail

distanceWeighting

protected int distanceWeighting

Whether the neighbors should be distance-weighted.

WEIGHT_NONE

public static final int WEIGHT_NONE

no weighting.

See Also:

Constant Field Values

WEIGHT_INVERSE

public static final int WEIGHT_INVERSE

weight by 1/distance.

See Also:

Constant Field Values

WEIGHT_SIMILARITY

public static final int WEIGHT_SIMILARITY

weight by 1-distance.

See Also:

Constant Field Values

numOfNeighbors

protected int numOfNeighbors

Number of neighbors used in the k-nearest neighbor algorithm

lnn

protected LinearNNSearch lnn

Class implementing the brute force search algorithm for nearest neighbor search. Default value is true.

dfunc

protected DistanceFunction dfunc

Implementing Euclidean distance (or similarity) function.

train

protected Instances train

The training instances

Constructor Detail

MultiLabelKNN

public MultiLabelKNN()

The default constructor

MultiLabelKNN

public MultiLabelKNN(int numOfNeighbors)

Initializes the number of neighbors

Parameters:

numOfNeighbors - the number of neighbors

Method Detail

setDfunc

public void setDfunc(DistanceFunction dfunc)

Sets a distance function

Parameters:

dfunc - the distance function

buildInternal

protected void buildInternal(MultiLabelInstances trainSet)
                      throws Exception

Description copied from class: MultiLabelLearnerBase

Learner specific implementation of building the model from MultiLabelInstances training data set. This method is called from MultiLabelLearnerBase.build(MultiLabelInstances) method, where behavior common across all learners is applied.

Specified by:

buildInternal in class MultiLabelLearnerBase

Parameters:

trainSet - the training data set

Throws:

Exception - if learner model was not created successfully

isUpdatable

public boolean isUpdatable()

Description copied from interface: MultiLabelLearner

Returns value indicating if learner is updatable, so if learner is able to perform on-line learning. The fact if learner is updatable or not influences the behavior of MultiLabelLearner.build(MultiLabelInstances) method.

When false is returned, each call of the MultiLabelLearner.build(MultiLabelInstances) will initialize the learner from the scratch, removing any potential knowledge built by previously entered training data. When true is returned, then on the first call of the MultiLabelLearner.build(MultiLabelInstances) the learner is initialized with the passed training data. All other calls contribute to the existing learner's model with new data.

Specified by:

isUpdatable in interface MultiLabelLearner

Overrides:

isUpdatable in class MultiLabelLearnerBase

Returns:

true if learner is updatable (on-line), false otherwise.

setDistanceWeighting

public void setDistanceWeighting(int distanceWeighting)

Parameters:

distanceWeighting - the distanceWeighting to set