Enhancing Cybersecurity using Machine Learning Strategies for Phishing Defense
C Jamunadevi, S. Pandikumar, Navile Nageshwara Naveen, S. Prasath, S Dhivyashree · 2024
Cybersecurity experts continue to be concerned about phishing attacks because it causes a real risk to people, businesses, and economies all around the world. Phishing attacks cause a critical threat to global cybersecurity, necessitating effective classification methods for phishing URL operations. So, this work aims to categorize phishing URL operations into distinct types: deceptive, spear, whaling, and pharming. Leveraging machine learning techniques including logistic regression, support vector machine, Naive Bayes, decision tree, K-NN, and Perceptron, evaluate their efficacy in accurately identifying and classifying phishing attacks. The findings underscore the importance of these models in providing cybersecurity professionals with actionable insights into the diverse tactics employed by malicious actors. Among the evaluated models, logistic regression emerges as the preferred choice, demonstrating superior performance in accurately categorizing phishing URLs. This work contributes to enhancing cybersecurity defenses by offering a comprehensive framework for understanding and mitigating the impact of phishing attacks on digital ecosystems.