Package com.actelion.research.chem

Class Clusterer<T>

java.lang.Object

com.actelion.research.calc.DataProcessor

com.actelion.research.chem.Clusterer<T>

public class Clusterer<T>
extends DataProcessor

Constructor Summary

Constructors

Constructor	Description
Clusterer(DescriptorHandler<T,?> descriptorHandler, T[] descriptor)

Method Summary

Modifier and Type	Method	Description
void	cluster(double similarityLimit, int clusterCountLimit)	Defines the criteria for stopping the clustering.
int	getClusterCount()
int	getClusterNo(int compound)
float	getSimilarity(int index1, int index2)
boolean	isRepresentative(int compound)
void	regenerateClusterNos()	Renumber cluster numbers starting at 1 to eliminate unused numbers.

Methods inherited from class com.actelion.research.calc.DataProcessor

addProgressListener, removeProgressListener, setThreadMaster, setVerbose, startProgress, stopProgress, threadMustDie, updateProgress

Methods inherited from class java.lang.Object

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

Constructor Detail

Clusterer

public Clusterer(DescriptorHandler<T,?> descriptorHandler,
                 T[] descriptor)

Method Detail

cluster

public void cluster(double similarityLimit,
                    int clusterCountLimit)

Defines the criteria for stopping the clustering. At least one of the two limits must be in the applicable valid range.

Parameters:

similarityLimit - >0...<=1.0 or 0.0 if not applied

clusterCountLimit - >=2...objectCount or -1 if not applied

isRepresentative

public boolean isRepresentative(int compound)

getClusterNo

public int getClusterNo(int compound)

getClusterCount

public int getClusterCount()

regenerateClusterNos

public void regenerateClusterNos()

Renumber cluster numbers starting at 1 to eliminate unused numbers.

getSimilarity

public float getSimilarity(int index1,
                           int index2)