Pages: [1]
Author Topic: Joining two examplesets by Clusters.  (Read 1117 times)
Full Member
Posts: 141

« on: March 23, 2009, 02:15:26 PM »


I have an exampleset on which kmeans is run and a variable 'Cluster' giving the cluster groupings in the data. There is also an ID variable in the same data.

I have an another exampleset, giving the cluster centroids. This data also has a variable called cluster, having the values same as the first exampleset, but each cluster appears only once. If there are 3 clusters, then there are only three examples here. This exampleset does NOT have an Id variable.

In both the examplesets, the type of the 'cluster' variable is regular.

Now, I want to join the above datasets by the 'cluster' variable. But the error i get is:"Input exampleset does not have an ID attribute". But i want to join the data by 'cluster' variable. How do i do it? Do i have to convert the regular type cluster variables to 'Id' variable in both the examplesets?

Thank you very much for your help,

Ingo Mierswa
Hero Member
Posts: 1232

« Reply #1 on: March 25, 2009, 11:08:21 AM »

If I understand you correctly: yes.

Why did you not simply try it out?


Did you try our new Marketplace? Upload or download new Extensions, add comments, and organize your operators. Have a look at
Pages: [1]
Jump to: