• In scikit-learn, CountVectorizer() counts the number of times a word shows up in the document and uses that value as its weight. You can also use TfidfVectorizer(), where the weight assigned to each token depends on both its frequency in a document and how often the term recurs in the entire corpus. You can find more on TfidfVectorizer at the following link: https://bit.ly/2sJCoVN.
• The scikit-learn documentation on the Naive Bayes classifier for multivariate Bernoulli models: https://bit.ly/2y3fASv.
• The scikit-learn documentation on the Naive Bayes classifier for multinomial models: https://bit.ly/2P4Ohic.