Machine Learning Classifier Performance Comparison for Phishing Detection

Abstract

Technology has advanced at a remarkable rate in recent decades, making communication simpler. Emails are the most effective method for both casual and formal conversations when compared to other forms of communication. Emails are a common form of communication for both business and personal purposes. Unfortunately, emails are also used to annoy internet users by sending viruses, spam, and ads. Spam emails are those sent by some undesirable users, also referred to as spammers. Some individuals misuse this kind of communication by sending spam emails that contain links to specific URLs for users to click on. Spam generates a number of issues, some of which may result in financial losses. Phishing is a method for attempting to obtain sensitive data through fraudulent email or website solicitation. This study compares the effectiveness of different machine learning algorithms in identifying spam or phishing emails. Different performance criteria were taken into account when evaluating the models, and the outcomes were compared.