Enhancing Cybersecurity through Genetic Algorithm-Based Phishing Website Detection

Cybercriminals design false webpages in an attempt to trick people into disclosing private information, such as passwords and credit card numbers. Strong cybersecurity measures are desperately needed, as phishing attacks are becoming more and more sophisticated. To improve cybersecurity, this study investigates the use of genetic algorithms for phishing website detection. We have used three real-world phishing website datasets to implement our work. Additionally, a thorough methodology that includes feature selection, data preprocessing, and the iterative process of the genetic algorithm is employed. This work proposes a hybrid-based method for phishing identification based on the website's URL that combines genetic and machine learning methods. The algorithm's effectiveness is demonstrated by the results, which also show its adaptability, relative advantages over conventional methods, and promising generalizability. By shedding light on the useful applications of genetic algorithms for phishing detection, the study advances the field of cybersecurity. The final model proposed in this work has a 99.99% accuracy rate on the test set.