Machine Learning Algorithms for Phishing Email Detection

Internet users are seriously endangered by phishing emails and keeping digital communication secure depends on their detection. The development of phishing strategies has necessitated continual research into increasingly sophisticated phishing email detection methods. Automatically spotting phishing emails has been demonstrated to be a powerful tool via machine learning (ML). In this study, we thoroughly examine current ML-based classifiers for accurately detecting phishing email. First, we employ a real-world dataset from Kaggle that has actual ratios of authentic and phishing emails. Then, Exploratory Data Analysis (EDA) is performed to understand the dataset better and identify obvious errors and outliers to help the detection process. Several ML methods, including Naive Bayes (NB), Support Vector Machine (SVM), Random Forest (RF), Decision Tree (DT), and K-Nearest Neighbors (KNN), are trained on the dataset. The receiver operating characteristic curve (ROC) and Area under the ROC curve (AUC) are the measures used to assess the models' performance. In addition, the time taken to train and test the model will be experimented to conclude the algorithm's efficiency in real-time. The findings show that random forest and decision tree classifier has the highest precision, recall, and f1-score, which concludes that these classifiers efficiently identify many positive instances. This study sheds important light on using ML for phishing email detection