An Ideal Machine Learning Method that uses URL Information to Identify Phishing Attempts

More websites have started gathering personal data in recent years for a variety of purposes, including banking, internet access, government services, and more. All personal information, including Aadhar, PAN, date of birth, phone number, etc., must be provided by the public. Through URL modification, personal data like Aadhar, PAN, date of birth, and phone number can be utilized for phishing attacks. Terrorists can make SIM cards using their names and other information since the public has lost money. It is of great importance to international security. In order to get around this, we employ the suggested techniques to identify phishing attempts by looking at factors such as the URL, site traffic, customer reviews, and the length of the business. These parameters allow the suggested optimal machine learningbased algorithm (OmLA), which was used to analyze historical URL data, to determine whether or not a given URL is phishing- or non-phishing-based. The suggested approach beats traditional techniques like random forest, support vector machine (SVM), and genetic algorithms by 8%, 18%, and 23%, respectively, in terms of accuracy, according to simulation and performance analysis. Furthermore, it excels in response times of 0.45%, 0.56%, and 0.62%, and achieves detection times of 0.2%, 0.6%, and 0.9%, respectively.