weka.core.neighboursearch.kdtrees

Class SlidingMidPointOfWidestSide

java.lang.Object

weka.core.neighboursearch.kdtrees.KDTreeNodeSplitter

weka.core.neighboursearch.kdtrees.SlidingMidPointOfWidestSide

All Implemented Interfaces:

java.io.Serializable, OptionHandler, RevisionHandler, TechnicalInformationHandler

public class SlidingMidPointOfWidestSide
extends KDTreeNodeSplitter
implements TechnicalInformationHandler

The class that splits a node into two based on the midpoint value of the dimension in which the node's rectangle is widest. If after splitting one side is empty then it is slided towards the non-empty side until there is at least one point on the empty side.

For more information see also:

David M. Mount (2006). ANN Programming Manual. College Park, MD, USA.

BibTeX:

 @manual{Mount2006,
    address = {College Park, MD, USA},
    author = {David M. Mount},
    organization = {Department of Computer Science, University of Maryland},
    title = {ANN Programming Manual},
    year = {2006},
    HTTP = {Available from http://www.cs.umd.edu/\~mount/ANN/}
 }
 

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
protected static double	ERR The floating point error to tolerate in finding the widest rectangular side.

Fields inherited from class weka.core.neighboursearch.kdtrees.KDTreeNodeSplitter

m_EuclideanDistance, m_Instances, m_InstList, m_NormalizeNodeWidth, MAX, MIN, WIDTH

Constructor Summary

Constructors

Constructor and Description
SlidingMidPointOfWidestSide()

Method Summary

Modifier and Type	Method and Description
java.lang.String	getRevision() Returns the revision string.
TechnicalInformation	getTechnicalInformation() Returns an instance of a TechnicalInformation object, containing detailed information about the technical background of this class, e.g., paper reference or book this class is based on.
java.lang.String	globalInfo() Returns a string describing this nearest neighbour search algorithm.
protected int	rearrangePoints(int[] indices, int startidx, int endidx, int splitDim, double splitVal) Re-arranges the indices array such that the points <= to the splitVal are on the left of the array and those > the splitVal are on the right.
void	splitNode(KDTreeNode node, int numNodesCreated, double[][] nodeRanges, double[][] universe) Splits a node into two based on the midpoint value of the dimension in which the node's rectangle is widest.

Methods inherited from class weka.core.neighboursearch.kdtrees.KDTreeNodeSplitter

correctlyInitialized, getOptions, listOptions, setEuclideanDistanceFunction, setInstanceList, setInstances, setNodeWidthNormalization, setOptions, widestDim

Methods inherited from class java.lang.Object

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

Field Detail

ERR

protected static double ERR

The floating point error to tolerate in finding the widest rectangular side.

Constructor Detail

SlidingMidPointOfWidestSide

public SlidingMidPointOfWidestSide()

Method Detail

globalInfo

public java.lang.String globalInfo()

Returns a string describing this nearest neighbour search algorithm.

Returns:

a description of the algorithm for displaying in the explorer/experimenter gui

getTechnicalInformation

public TechnicalInformation getTechnicalInformation()

Returns an instance of a TechnicalInformation object, containing detailed information about the technical background of this class, e.g., paper reference or book this class is based on.

Specified by:

getTechnicalInformation in interface TechnicalInformationHandler

Returns:

the technical information about this class

splitNode

public void splitNode(KDTreeNode node,
                      int numNodesCreated,
                      double[][] nodeRanges,
                      double[][] universe)
               throws java.lang.Exception

Splits a node into two based on the midpoint value of the dimension in which the node's rectangle is widest. If after splitting one side is empty then it is slided towards the non-empty side until there is at least one point on the empty side. The two nodes created after the whole splitting are correctly initialised. And, node.left and node.right are set appropriately.

Specified by:

splitNode in class KDTreeNodeSplitter

Parameters:

node - The node to split.

numNodesCreated - The number of nodes that so far have been created for the tree, so that the newly created nodes are assigned correct/meaningful node numbers/ids.

nodeRanges - The attributes' range for the points inside the node that is to be split.

universe - The attributes' range for the whole point-space.

Throws:

java.lang.Exception - If there is some problem in splitting the given node.

rearrangePoints

protected int rearrangePoints(int[] indices,
                              int startidx,
                              int endidx,
                              int splitDim,
                              double splitVal)

Re-arranges the indices array such that the points <= to the splitVal are on the left of the array and those > the splitVal are on the right.

Parameters:

indices - The master index array.

startidx - The begining index of portion of indices that needs re-arranging.

endidx - The end index of portion of indices that needs re-arranging.

splitDim - The split dimension/attribute.

splitVal - The split value.

Returns:

The startIdx of the points > the splitVal (the points belonging to the right child of the node).

getRevision

public java.lang.String getRevision()

Returns the revision string.

Specified by:

getRevision in interface RevisionHandler

Overrides:

getRevision in class KDTreeNodeSplitter

Returns:

the revision