Phishing URL Detection using Artificial Neural Network

Abstract

URL phishing is a developing problem in which fraudsters create fake websites to entice victims into giving up vital information. These bogus websites frequently resemble the actual thing, so looking for telltale signals might help protect you from URL phishing. Organizations can reduce their danger by training users and implementing automatic email-screening measures. We provided a method to categorize URLs as real or phishing URLs in this study. The data was collected and the selective features were extracted from the URLs. We constructed a dataset with a mix of phishing and authentic URLs after extracting features based on three criteria. From a total of 10,000 URLs, we were able to extract 18 features, with 5000 phishing and 5000 genuine URLs. Naley Decision Tree, Random Forest, Support Vector Machine, and Artificial Neural Network were used as machine learning models. ANN has a maximum accuracy of 84.35 percent on these models.