Regresar

Phishing attack detection: A solution based on the typical machine learning modeling cycle

Abstract:

The aim of the current study has been the design and development of a model for detecting Phishing attacks using supervised Machine Learning techniques. Thus, we conducted a literature review to identify the features of infected emails by phishing. As a result, a combination of the Naive Bayes and Decision tree algorithms has been constructed using the typical cycle of Machine Learning (ML) modeling. The main tools used have been Jupiter framework and Python. The proof of concept has been performed in a controlled environment. The infected emails has been obtained using PhishTank. Finally, to yield the higher level of accuracy of phishing detection, the validation of results was accomplished using the most accepted algorithms in the scientific field such as ML Random Forest, Logistic Regression and Fictitious Classifier, according to our literature review.