Hi everyone, I am very new to this field and I don’t have a well background. I am using Text Blob’s built-in classifier for multi-classes text classification. I think Text Blob is using NLTK classifier. I am using prob_classify to find the probability of each class. Everything is working fine, but when I am trying to manually calculate the frequency table and then use the Naive Bayes formula to get the probability of each class I get different numbers from what I get from prob_Classify function.

P(x|y) = ( P(y|x) P(x) ) / P(y)

I used the above formula and then I tried to add log to the probabilities but my result still doesn’t match the Text Blob classifier. I’ve read the code of the prob_classify function. But, unfortunately I didn’t get the idea.