Package elki.datasource.filter.transform

Class LinearDiscriminantAnalysisFilter<V extends elki.data.NumberVector>

java.lang.Object

elki.datasource.filter.transform.AbstractSupervisedProjectionVectorFilter<V>

elki.datasource.filter.transform.LinearDiscriminantAnalysisFilter<V>

Type Parameters:

V - Vector type

All Implemented Interfaces:

elki.datasource.filter.ObjectFilter

@Alias("lda")
@Reference(authors="R. A. Fisher",
           title="The use of multiple measurements in taxonomic problems",
           booktitle="Annals of Eugenics 7.2",
           url="https://doi.org/10.1111/j.1469-1809.1936.tb02137.x",
           bibkey="doi:10.1111/j.1469-1809.1936.tb02137.x")
public class LinearDiscriminantAnalysisFilter<V extends elki.data.NumberVector>
extends AbstractSupervisedProjectionVectorFilter<V>

Linear Discriminant Analysis (LDA) / Fisher's linear discriminant.

Reference:

R. A. Fisher
The use of multiple measurements in taxonomic problems
Annals of Eugenics 7.2

Since:

0.6.0

Author:

Angela Peng, Erich Schubert

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static class	LinearDiscriminantAnalysisFilter.Par<V extends elki.data.NumberVector>	Parameterization class.

Field Summary

Fields

Modifier and Type	Field	Description
private static elki.logging.Logging	LOG	Class logger.

Fields inherited from class elki.datasource.filter.transform.AbstractSupervisedProjectionVectorFilter

tdim

Constructor Summary

Constructors

Constructor	Description
LinearDiscriminantAnalysisFilter(int projdimension)	Constructor.

Method Summary

Modifier and Type	Method	Description
protected java.util.List<elki.math.linearalgebra.Centroid>	computeCentroids(int dim, java.util.List<V> vectorcolumn, java.util.List<elki.data.ClassLabel> keys, java.util.Map<elki.data.ClassLabel,it.unimi.dsi.fastutil.ints.IntList> classes)	Compute the centroid for each class.
protected double[][]	computeProjectionMatrix(java.util.List<V> vectorcolumn, java.util.List<? extends elki.data.ClassLabel> classcolumn, int dim)	computes the projection matrix
protected elki.logging.Logging	getLogger()	Class logger.

Methods inherited from class elki.datasource.filter.transform.AbstractSupervisedProjectionVectorFilter

convertedType, filter, partition

Methods inherited from class java.lang.Object

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

Field Detail

LOG

private static final elki.logging.Logging LOG

Class logger.

Constructor Detail

LinearDiscriminantAnalysisFilter

public LinearDiscriminantAnalysisFilter(int projdimension)

Constructor.

Parameters:

projdimension - Projection dimensionality.

Method Detail

computeProjectionMatrix

protected double[][] computeProjectionMatrix(java.util.List<V> vectorcolumn,
                                             java.util.List<? extends elki.data.ClassLabel> classcolumn,
                                             int dim)

Description copied from class: AbstractSupervisedProjectionVectorFilter

computes the projection matrix

Specified by:

computeProjectionMatrix in class AbstractSupervisedProjectionVectorFilter<V extends elki.data.NumberVector>

Parameters:

vectorcolumn - Vectors

classcolumn - Class information

dim - Dimensionality Dimensionality

Returns:

Projection matrix

computeCentroids

protected java.util.List<elki.math.linearalgebra.Centroid> computeCentroids(int dim,
                                                                            java.util.List<V> vectorcolumn,
                                                                            java.util.List<elki.data.ClassLabel> keys,
                                                                            java.util.Map<elki.data.ClassLabel,it.unimi.dsi.fastutil.ints.IntList> classes)

Compute the centroid for each class.

Parameters:

dim - Dimensionality

vectorcolumn - Vector column

keys - Key index

classes - Classes

Returns:

Centroids for each class.

getLogger

protected elki.logging.Logging getLogger()

Class logger.

Specified by:

getLogger in class AbstractSupervisedProjectionVectorFilter<V extends elki.data.NumberVector>

Returns:

Logger