Phishing URL Classification Analysis Using ANN Algorithm

Over the years in social engineering, phishing URLs became a significant threat for internet users as most cyber-crime or attacks redirect to individual target users by sending a malicious or crafted URL which allows the attacker to capture sensitive pieces of information of the victim users. According to this research, most of the existing phishing detection tools showed an overall accuracy level of 70% to 92.52%. This paper introduced a random forest (RF) and artificial neural network(ANN) based machine learning (ML) algorithmic classification models for detecting phishing URLs effectively with higher accuracy. The experimental results showed that the proposed RF and ANN models could classify phishing URLs' legitimacy labels with an efficiency rate of up to 99%, much more significant than traditional ways. In this paper, this comparison's algorithmic learning models demonstrate that the proposed ANN classification holds a competent accuracy label of 98.72%.