The thesis explores machine learning algorithms in cybersecurity, especially from the detection and prevention points of view against phishing attacks. Several classification models like Naive Bayes, Logistic Regression, SGD Classifier, Decision Tree, Random Forest, and XGBoost are trained and tested using a real-world email dataset. These machine learning models were compared using performance metrics such as accuracy and F1-score to analyze the strengths and weaknesses of each model. All the implementation details and performance results are on my Github repository.