This document introduces Naive Bayes classification. It discusses using Bayes' rule to calculate the probability of an email being spam given the presence of a word. An example is worked out classifying an email as spam or ham based on the word "meeting". The document then expands on this to consider multiple words using a naive Bayes model that treats words as independent predictors. It notes that wrangling data is important and discusses extracting and classifying articles from the NYT using Naive Bayes as an example.