Package org.aksw.limes.core.ml.algorithm

Class ACoreMLAlgorithm

java.lang.Object

org.aksw.limes.core.ml.algorithm.ACoreMLAlgorithm

Direct Known Subclasses:

ACIDS, AWombat, Dragon, Eagle, LinearEuclid

public abstract class ACoreMLAlgorithm
extends Object

Version:

Jul 15, 2016

Author:

Mohamed Sherif (sherif@informatik.uni-leipzig.de)

Field Summary

Fields

Modifier and Type	Field	Description
protected Configuration	configuration
protected List<LearningParameter>	learningParameters
protected static org.slf4j.Logger	logger
protected ACache	sourceCache
protected ACache	targetCache

Constructor Summary

Constructors

Constructor	Description
ACoreMLAlgorithm()

Method Summary

Modifier and Type	Method	Description
protected abstract MLResults	activeLearn()	Learning method for supervised active core ML algorithm implementations Normally, it is used as a first step to initialize the ML model before going through the active learning process
protected abstract MLResults	activeLearn(AMapping oracleMapping)	Learning method for supervised active core ML algorithm implementations.
Configuration	getConfiguration()
protected abstract String	getName()	Name of the core ML algorithm.
protected abstract AMapping	getNextExamples(int size)	Get a set of examples to be added to the mapping.
protected Object	getParameter(String name)	Get parameter by name.
protected List<LearningParameter>	getParameters()
ACache	getSourceCache()
ACache	getTargetCache()
protected void	init(List<LearningParameter> learningParameters, ACache sourceCache, ACache targetCache)	Initialize the core ML algorithm.
protected abstract MLResults	learn(PseudoFMeasure pfm)	Learning method for unsupervised core ML algorithm implementations.
protected abstract MLResults	learn(AMapping trainingData)	Learning method for supervised core ML algorithm implementations, where the confidence values for each pair in the trainingData determine its truth degree.
protected abstract AMapping	predict(ACache source, ACache target, MLResults mlModel)	Predict/generate links from source to target based on mlModel.
void	setConfiguration(Configuration configuration)
abstract void	setDefaultParameters()	Set default ACoreMLAlgorithm parameters values
void	setParameter(String par, Object val)
protected abstract boolean	supports(MLImplementationType mlType)	Check whether the mlType is supported.

Methods inherited from class java.lang.Object

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

Field Detail

logger

protected static org.slf4j.Logger logger

learningParameters

protected List<LearningParameter> learningParameters

sourceCache

protected ACache sourceCache

targetCache

protected ACache targetCache

configuration

protected Configuration configuration

Constructor Detail

ACoreMLAlgorithm

public ACoreMLAlgorithm()

Method Detail

getConfiguration

public Configuration getConfiguration()

Returns:

the configuration

setConfiguration

public void setConfiguration(Configuration configuration)

Parameters:

configuration - the configuration

getName

protected abstract String getName()

Name of the core ML algorithm.

Returns:

Name of the core ML algorithm.

getParameters

protected List<LearningParameter> getParameters()

Returns:

current core ML algorithm parameters and their default values

setDefaultParameters

public abstract void setDefaultParameters()

Set default ACoreMLAlgorithm parameters values

init

protected void init(List<LearningParameter> learningParameters,
                    ACache sourceCache,
                    ACache targetCache)

Initialize the core ML algorithm.

Parameters:

learningParameters - learning parameters

sourceCache - the source cache

targetCache - the target cache

learn

protected abstract MLResults learn(AMapping trainingData)
                            throws UnsupportedMLImplementationException

Learning method for supervised core ML algorithm implementations, where the confidence values for each pair in the trainingData determine its truth degree.

Parameters:

trainingData - used for learning

Returns:

wrap with results

Throws:

UnsupportedMLImplementationException - if ML implementation is not supported

learn

protected abstract MLResults learn(PseudoFMeasure pfm)
                            throws UnsupportedMLImplementationException

Learning method for unsupervised core ML algorithm implementations.

Parameters:

pfm - pseudo F-measure for unsupervised learning

Returns:

wrap with results

Throws:

UnsupportedMLImplementationException - if ML implementation is not supported

predict

protected abstract AMapping predict(ACache source,
                                    ACache target,
                                    MLResults mlModel)

Predict/generate links from source to target based on mlModel.

Parameters:

source - Cache

target - Cache

mlModel - result of training phase

Returns:

the mapping from source to target

supports

protected abstract boolean supports(MLImplementationType mlType)

Check whether the mlType is supported.

Parameters:

mlType - machine learning implementation type

Returns:

a boolean value

getNextExamples

protected abstract AMapping getNextExamples(int size)
                                     throws UnsupportedMLImplementationException

Get a set of examples to be added to the mapping.

Parameters:

size - of the examples

Returns:

the mapping

Throws:

UnsupportedMLImplementationException - Exception

activeLearn

protected abstract MLResults activeLearn(AMapping oracleMapping)
                                  throws UnsupportedMLImplementationException

Learning method for supervised active core ML algorithm implementations.

Parameters:

oracleMapping - mapping from the oracle

Returns:

wrap with results

Throws:

UnsupportedMLImplementationException - if ML implementation is not supported

activeLearn

protected abstract MLResults activeLearn()
                                  throws UnsupportedMLImplementationException

Learning method for supervised active core ML algorithm implementations Normally, it is used as a first step to initialize the ML model before going through the active learning process

Returns:

wrap with results

Throws:

UnsupportedMLImplementationException - if ML implementation is not supported

getParameter

protected Object getParameter(String name)

Get parameter by name.

Parameters:

name - parameter name

Returns:

the parameter as Object

Throws:

NoSuchParameterException - if parameter is not exists

setParameter

public void setParameter(String par,
                         Object val)

Parameters:

par - parameter name

val - parameter value

getSourceCache

public ACache getSourceCache()

Returns:

the source cache

getTargetCache

public ACache getTargetCache()

Returns:

the target cache