TF(phrase in a doc)= Variety of occurrences of that word in doc / Number of text in document and predicts according to which joint distribution has the highest probability. The naive assumption during the Naive Bayes product is that the person words are impartial. So: The phrase index is a https://crmsystems80016.blogaritma.com/35043460/how-support-systems-can-save-you-time-stress-and-money