In this assignment, we were asked to implement Naïve Bayes document classifier and apply it to the 20 newsgroups dataset. In this dataset, each document is a posting that was made to one of 20 different newsgroups. Multinomial event modeling is used to estimate the parameters and Laplace smoothing is applied in order to avoid zero probabilities. %74.8 accuracy obtanied after tuning. Detailed report on accuracy and Confusion Matrix can be found in the report.