I am working on a text classification problem, and plan on using Naive Bayes based model.
In which cases should I consider using Multinomial Naive Bayes?
For Text classification Multinomial Naive bayes is a good choice to start with.
We can start with MNB, when
1. There is no dependence between the predictors
2. Good if class is not unblanced
(Here is an approach to handle unbalanced data with MNB )
I would start with bag of words and then TFIDF vectors to check the model performance before moving further.