Multinomial Naive Bayes discrete or continuous

#1

In this video, it is told that multinomial feature vectors will contain discrete values.
But after using tf idf, how will the vectors have discrete values? they will be continuous values (weights) right?
A bit of confusion.

#2

Hey @muditarya31, so we are talking about the feature values before any processing. This means we are seeing only the feature values, so in that terms we are saying multinomial deals with discrete values features. We use tf-idf after having those values features in calculating conditional probabilites. You can also think it like this way, you can’t apply multinomial on a dataset having feature like X ( which can take any real values from 0 to 100). X can take 0.004 or 99.55656 anything. So do you think you will be able to apply multinomial in this case ?. ( Definitely not, but yes gaussian naive bayes can handle this as well).

hope this resolved your doubt. :blush:

#3

I hope I’ve cleared your doubt. I ask you to please rate your experience here
Your feedback is very important. It helps us improve our platform and hence provide you
the learning experience you deserve.

On the off chance, you still have some questions or not find the answers satisfactory, you may reopen
the doubt.