Phishing URL Detection Using Machine Learning Classification Algorithms

Jeremiah Paul Richard; Matthias  Daniel

Abstract

Phishing attack is utilized to get the data like username, secret phrase, financial balance subtleties, and credit card details. Today, is the most well-known cybercrime. Phishing assaults additionally influence the web-based installment area monetary organization, document facilitating or distributed storage, and numerous others. Phishing assault generally focuses to these Web locales which are connected with the internet-based payment area and Web mail. To stop phishing attacks, a variety of methods are employed, including blacklists, heuristics, and visual similarity. The proposed model in this research, however, is a combination of logistic regression (LR) and decision tree (DT) with some variable parameters related to the design and training of classifiers used by data mining techniques related to real-world issues using Python programming language (Spyder IDE) used in developing the model with Sklearn built-in data source library. The system was successfully tested according to the design specification using logistic regression and Decision Tree models. This resulted to 86.33 and 87.62% accuracy level in comparison with the existing system that has 81.42% accuracy rate.