Code covered by the BSD License  

Highlights from
Clustering results measurement

Be the first to rate this file! 22 Downloads (last 30 days) File Size: 1.61 KB File ID: #32197

Clustering results measurement

by

 

12 Jul 2011 (Updated )

Measure percentage of Accuracy and the Rand index of clustering results

| Watch this File

File Information
Description

Measure percentage of Accuracy and the Rand index of clustering results
The number of class must equal to the number cluster

Output
 Acc = Accuracy of clustering results
 rand_index = Rand's Index, measure an agreement of the clustering results
 match = 2xk matrix which are the best match of the Target and clustering results

Input
 T = 1xn target index
 idx =1xn matrix of the clustering results

 EX:
 X=[randn(200,2);randn(200,2)+6,;[randn(200,1)+12,randn(200,1)]]; T=[ones(200,1);ones(200,1).*2;ones(200,1).*3];
 idx=kmeans(X,3,'emptyaction','singleton','Replicates',5);
  [Acc,rand_index,match]=Acc_measure(T,idx)

MATLAB release MATLAB 7 (R14)
Tags for This File   Please login to tag files.
Please login to add a comment or rating.
Comments and Ratings (1)
30 Jan 2014 Pedro

there is a bug in your code since AccMeasure([1 1 1],[2 2 1]) and AccMeasure([2 2 2],[1 1 2]) give different accuracies

Updates
11 Aug 2011

correct the example code

Contact us