The purpose of the project is to implement two machine learning algorithms, Naive Bayes and ID3, and demonstrate algorithm learning capabilities using a set of data (enron). There are 3 different enrons, which contain both spam and ham messages.
Initially, the messages are divided into two individual sets, the train (80%) and the test(20%) with the help of the islice method. After the algoriths have been trained on the train set, their performance is checked on both the train and the test set, which was not taken into account during the training process.