jsat.datatransform.visualization

Class Isomap

java.lang.Object

jsat.datatransform.visualization.Isomap

All Implemented Interfaces:

Serializable, Cloneable, VisualizationTransform

public class Isomap
extends Object
implements VisualizationTransform

Isomap is an extension of MDS. It uses a geodesic distance made from a nearest neighbor search of all the points in the data set. This implementation also includes the extension C-Isomap, which further weights distances by density.

Note, that Isomap normally will fail on some datasets when two or more regions can not be connected in the induced neighbor graph. While increasing the number of neighbors considered will eventually resolve this problem, the separated groups may be desirable in practice. This implementation includes a non-standard addition that will forcibly connect such isolated regions with very large values, hoping to preserve the farther distances in the given dataset while maintaining local structure.

See:

• Tenenbaum, J. B., Silva, V. De, & Langford, J. C. (2000). A Global Geometric Framework for Nonlinear Dimensionality Reduction. Science, 290, 2319–2323. doi:10.1126/science.290.5500.2319
• De Silva, V., & Tenenbaum, J. B. (2003). Global Versus Local Methods in Nonlinear Dimensionality Reduction. In Advances in Neural Information Processing Systems 15 (pp. 705–712). MIT Press. Retrieved from here

Author:

Edward Raff

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
Isomap()
Isomap(int searchNeighbors)
Isomap(int searchNeighbors, boolean c_isomap)

Method Summary

Modifier and Type	Method and Description
int	getNeighbors()
int	getTargetDimension()
boolean	isCIsomap()
void	setCIsomap(boolean c_isomap) Controls whether the C-Isomap extension is used.
void	setNeighbors(int searchNeighbors) Set the number of neighbors to consider for the initial graph in Isomap
boolean	setTargetDimension(int target) Sets the target dimension to embed new dataset to.
<Type extends DataSet> Type	transform(DataSet<Type> d) Transforms the given data set, returning a dataset of the same type.
<Type extends DataSet> Type	transform(DataSet<Type> d, ExecutorService ex) Transforms the given data set, returning a dataset of the same type.

Methods inherited from class java.lang.Object

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

Constructor Detail

Isomap

public Isomap()

Isomap

public Isomap(int searchNeighbors)

Parameters:

searchNeighbors - the number of nearest neighbors to consider

Isomap

public Isomap(int searchNeighbors,
              boolean c_isomap)

Parameters:

searchNeighbors - the number of nearest neighbors to consider

c_isomap - true to use the C-Isomap extension, false for normal Isomap.

Method Detail

setNeighbors

public void setNeighbors(int searchNeighbors)

Set the number of neighbors to consider for the initial graph in Isomap

Parameters:

searchNeighbors - the number of nearest neighbors to consider

getNeighbors

public int getNeighbors()

Returns:

the number of neighbors used when creating the initial graph

setCIsomap

public void setCIsomap(boolean c_isomap)

Controls whether the C-Isomap extension is used. If set true, the initial distances will also be scaled based on the density of the region between the points. If false, normal Isomap will be used.

Parameters:

c_isomap - true to use the C-Isomap extension, false for normal Isomap.

isCIsomap

public boolean isCIsomap()

Returns:

true if the C-Isomap extension is in use, false for normal Isomap.

transform

public <Type extends DataSet> Type transform(DataSet<Type> d)

Description copied from interface: VisualizationTransform

Transforms the given data set, returning a dataset of the same type.

Specified by:

transform in interface VisualizationTransform

Type Parameters:

Type - the dataset type

Parameters:

d - the data set to transform

Returns:

the lower dimension dataset for visualization.

transform

public <Type extends DataSet> Type transform(DataSet<Type> d,
                                             ExecutorService ex)

Description copied from interface: VisualizationTransform

Transforms the given data set, returning a dataset of the same type.

Specified by:

transform in interface VisualizationTransform

Type Parameters:

Type - the dataset type

Parameters:

d - the data set to transform

ex - the source of threads for parallel computation

Returns:

the lower dimension dataset for visualization.

getTargetDimension

public int getTargetDimension()

Specified by:

getTargetDimension in interface VisualizationTransform

Returns:

the number of dimensions that a dataset will be embedded down to

setTargetDimension

public boolean setTargetDimension(int target)

Description copied from interface: VisualizationTransform

Sets the target dimension to embed new dataset to. Many visualization methods may only support a target of 2 or 3 dimensions, or only one of those options. For that reason a boolean value will be returned indicating if the target size was acceptable. If not, no change to the object will occur.

Specified by:

setTargetDimension in interface VisualizationTransform

Parameters:

target - the new target dimension size when VisualizationTransform.transform(jsat.DataSet) is called.

Returns:

true if this transform supports that dimension and it was set, false if the target dimension is unsupported and the previous value will be used instead.