Text mining excel using clustering to obtain confusion matrix

Clusters are a form of unsupervised machine learning so it is not possible to generate a confusion matrix directly from clustering.  You would first need to turn the clusters into a label and then have another process to assign the clusters to compare the two outputs.  Or if you already have another existing label with the same number of categories as clusters, then you can use the Map Clusters on Labels operator to do this automatically and then use a normal Performance operator to generate the confusion matrix.

I built this model to classify the confusion matrix. I managed to get the confusion matrix but I don't know if it is the correct form. Does it make sense to you?

hi @brunonbrasil so based on that screenshot you are using a very old version of RapidMiner Studio. I would highly recommend updating to the most recent version (9.5.1).

Scott

The context I consider as a label, means the clusters that I obtain manually and compare with the clusters that I intend to obtain. The Receiver represents the data in sentences.