Modelo de machine learning en la detección de sitios web phishing

Currently the growth of phishing attacks is evident, this work aims to develop a model for detecting phishing websites, based on machine learning and taking into account the characteristics of the URL, the source code and the intelligence of the threats of the websites. A data set of 30 characteristics of 11055 websites is used, Random Forest, Extra Tree and Decision Tree models are trained, Random Forest being the chosen model, performance was evaluated with data from 2211 websites and an accuracy of 97.56% is obtained, which is higher compared to the results of other models in previous works.