-2

I am working on Bayes and Naive Bayes theorem. I am stuck in the condition when the probabilities are the same for both labels. How does Naive Bayes handle this scenario?

Sushil
  • 3
  • 1
  • 2
    I’m voting to close this question because it is not about programming as defined in the [help] but about ML theory & methodology - please see the intro & **NOTE** in the `machine-learning` [tag info](https://stackoverflow.com/tags/machine-learning/info). – desertnaut Nov 22 '20 at 19:08

1 Answers1

1

In case of equality you may arbitrarily set the sample to any class, exactly for that reason: the sample can belong to any of the classes with the same probability, no preference.

Now you only need to select the strategy: you can choose constantly one specific class, use apriory class distribution, choose in random or using round-robin approach. Most simple is to choose randomly.

igrinis
  • 12,398
  • 20
  • 45