procope.methods.clustering
Class MCLPerfomanceCalculator

java.lang.Object
  extended by procope.methods.clustering.MCLPerfomanceCalculator

public class MCLPerfomanceCalculator
extends Object

Calculates the quality of a clustering using a given scores network according to the method proposed in the PhD of Stijn van Dongen (for a reference see: MarkovClusterer).

Basically the performance of a clusterings gets higher the more high-scoring edges there are within the clusters. Efficiency values are between 0.0 and 1.0.

Note that it might not make very much sense to evaluate a clustering with a scores network it was not derived from.

Author:
Jan Krumsiek

Constructor Summary
MCLPerfomanceCalculator()
           
 
Method Summary
static double calculatePerformance(ProteinNetwork network, ComplexSet clustering)
          Calculate performance of the clustering using a given complex network.
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Constructor Detail

MCLPerfomanceCalculator

public MCLPerfomanceCalculator()
Method Detail

calculatePerformance

public static double calculatePerformance(ProteinNetwork network,
                                          ComplexSet clustering)
Calculate performance of the clustering using a given complex network. Uses a value of 2.0 for the r parameter.

Parameters:
network - scores network used for evaluation
clustering - clustering to be evaluated
Returns:
efficiency value between 0.0 and 1.0