The article introduces a new classifier called naive Bayes, which uses probabilities to predict if an instance belongs to a certain class. It shows how this classifier can be used to filter spam emails by comparing them to previously labeled emails. The method is based on basic probability principles and differs from classical Bayesian methods. The article explains how to build and use a naive Bayes classifier in R to predict the class of new data. A dataset of over 1,600 emails labeled as "ham" or "spam" is used for analysis, highlighting the strengths and weaknesses of the naive Bayes method.