ca.pfv.spmf.algorithms.clustering.hierarchical_clustering

Class AlgoHierarchicalClustering

java.lang.Object

ca.pfv.spmf.algorithms.clustering.hierarchical_clustering.AlgoHierarchicalClustering

public class AlgoHierarchicalClustering
extends java.lang.Object

This is an implementation of generic Hierarchical Clustering Algorithm as described in this webpage: http://home.dei.polimi.it/matteucc/Clustering/tutorial_html/hierarchical.html

This is a Hierarchical Clustering with a constant "threshold" that indicate the maximal distance between two clusters to group them. The algorithm stops when no cluster can be merged.

The distance between two clusters is calculated as the distance between the medians of the two clusters.

Constructor Summary

Constructors

Constructor and Description
AlgoHierarchicalClustering() Default constructor

Method Summary

Modifier and Type	Method and Description
void	printStatistics() Print statistics about the latest execution to System.out.
java.util.List<Cluster>	runAlgorithm(java.lang.String inputFile, double maxDistance) Run the algorithm.
void	saveToFile(java.lang.String output) Save the clusters to an output file

Methods inherited from class java.lang.Object

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

Constructor Detail

AlgoHierarchicalClustering

public AlgoHierarchicalClustering()

Default constructor

Method Detail

runAlgorithm

public java.util.List<Cluster> runAlgorithm(java.lang.String inputFile,
                                            double maxDistance)
                                     throws java.lang.NumberFormatException,
                                            java.io.IOException

Run the algorithm.

Parameters:

inputFile - an input file containing vectors of doubles

maxDistance - the maximum distance allowed for merging two clusters

Returns:

a list of Clusters

Throws:

java.io.IOException - exception if error while reading the file

java.lang.NumberFormatException

saveToFile

public void saveToFile(java.lang.String output)
                throws java.io.IOException

Save the clusters to an output file

Parameters:

output - the output file path

Throws:

java.io.IOException - exception if there is some writing error.

printStatistics

public void printStatistics()

Print statistics about the latest execution to System.out.