Phishing attacks represent a significant threat in the realm of cybercrime, characterized by the use of deceptive emails and websites to extract sensitive information from unsuspecting individuals. Traditional detection methods have become less effective against increasingly sophisticated phishing techniques, necessitating the adoption of advanced technologies such as machine learning (ML) for enhanced detection capabilities. This research paper presents a comparative analysis of various machine learning algorithms—Naive Bayes, Logistic Regression, SGD Classifier, XGBoost, Decision Tree, Random Forest, and MLPClassifier—focusing on their effectiveness in phishing email classification. By evaluating key performance metrics such as accuracy, validation accuracy, classification reports, and confusion matrixes, the study identifies the strengths and weaknesses of each algorithm. The results highlight the SGD Classifier and Logistic Regression as the most robust models, exhibiting high generalization capabilities suitable for practical applications. Conversely, models like XGBoost and Decision Tree showed possible overfitting, indicating a need for further tuning. The findings provide valuable insights for cybersecurity professionals, aiding in the selection and implementation of effective phishing detection systems. Future research should explore hybrid models and advanced feature extraction techniques to further enhance detection accuracy and robustness.