[http://bit.ly/NBayes] How can we distinguish spam from nonspam with a Naive Bayes classifier? We estimate the priors and multiple Bernoulli distributions for each class. Also learn how Naive Bayes can misclassify its own training examples.