The project includes a Jupyter notebook file named 'q1_test.ipynb' which contains a detailed analysis of the algorithms used. The data is split into a 4:1 ratio for training and testing. The notebook provides a comprehensive analysis of both the training and testing data. It includes an evaluation of errors and accuracies, with graphical representations of errors versus iterations for the logistic regression and perceptron algorithms. The 'spam_class_ass3.py' script uses the entire dataset to predict errors using logistic regression, Naive Bayes, and SVM algorithms. The script takes the majority vote from the algorithms' responses to predict whether an email is spam or not. The predictions are exported to a file named 'prediction.csv'. The 'feature.txt' file contains all tokenized words from the dataset, providing an overview of the words present in the dataset. The 'spam_class_ass3.py' script also provides final predictions on the emails stored in the 'test' folder. The emails in the 'test' folder are processed in numerical order, i.e., 'email1.txt', 'email2.txt', 'email3.txt', and so on.