1

I have 6 class labels, say: a, b, c, d, e, and f.

I am using 3 classification models: decision tree, random forest, and naive Bayes.

Can majority voting be applied this ensemble, say more than 50% votes for a decision?

Mark White
  • 10,252
vidhi
  • 11
  • 2

1 Answers1

0

In this case there is a possibility that no class gets more votes than the others. This can be solved/controlled by using weighted voting. Let one of the members of the ensemble have slightly higher weight, to let it decide in case of a tie.

Whether this will give good results in your case is an open question.

Jon Nordby
  • 1,502