Phishing URL detection using machine learning methods

• Published in: Advances in engineering software (1992) Vol. 173; p. 103288
• Main Authors: Ahammad, SK Hasane, Kale, Sunil D., Upadhye, Gopal D., Pande, Sandeep Dwarkanath, Babu, E Venkatesh, Dhumane, Amol V., Bahadur, Mr. Dilip Kumar Jang
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01-11-2022
• Subjects: Cybercrime; Light GBM; Machine learning; Malicious website; Phishing; URL; Light GBM; Cybercrime; Phishing; Malicious website; URL; Machine learning
• ISSN: 0965-9978
• DOI: 10.1016/j.advengsoft.2022.103288

•Phishing is among the most concerning issues in a constantly changing world.•The increasing use of the Internet has led to a new way of stealing data, known as cybercrime.•Cybercrime refers to stealing private information and violating privacy through computers. The primary technique used is phishing.•Phishing via URLs (Uniform Resource Locators) is one of the most common types, and its primary goal is to steal the data from the user when the user accesses the malicious website.•This work aims to provide a solution for detecting such websites with the help of machine learning algorithms.
Phishing is among the most concerning issues in a constantly changing world. The increasing use of the Internet has led to a new way of stealing data, known as cybercrime. Cybercrime refers to stealing private information and violating privacy through computers. The primary technique used is phishing. Phishing via URLs (Uniform Resource Locators) is one of the most common types, and its primary goal is to steal the data from the user when the user accesses the malicious website. Detecting a malicious URL is a significant challenge. This work aims to provide a solution for detecting such websites with the help of machine learning algorithms focused on the behaviors and qualities of the suggested URL. The web security community has created blacklisting services to identify malicious websites. A variety of methods, such as manual reporting, and site analysis heuristics are used to create these blacklists. Due to their recentness, lack of evaluation, or incorrect evaluation, many malicious websites inadvertently escape blacklisting. To create a machine learning model for detecting whether a URL is malicious or not, algorithms such as Random Forests, Decision Trees, Light GBM, Logistic Regression, and Support Vector Machine (SVM) are used. Extracting features is the first step, and applying the model is the next step.