Hi, i am dealing with text categorization problem. I need confidence/probabilities values generated by Naive Bayes but i found that its giving me binary confidences(1/0). I have 20 classes in my dataset where one class confidence is calculated as '1' and others as '0'. It happens same for all correctly classified instances and misclassified instances.
I am using Modeling->Classification And Regression -> Bayesian Modeling ->Naive Bayes
am i using the correct naive bayes and why is it only generating binary values as probabilities ?
Thanks alot in advance
you are using the correct operator. Either your problem is algorithmically very easy, or you have an error/misconfiguration in your process setup, or your data has other particularities.
Can you please post the XML of your process such that we can have a look at it?
Best regards,
Marius