A Novel Approach to Detect Phishing Attack Using Artificial Neural Networks Combined with Pharming Detection

Phishing is the most common yet a major cyber crime. In this fraudulent practice, the perpetrator sends an e-mail to the target posing as a legitimate organization. This email contains a URL link to the phishing website which the user is prompted to visit and is induced to reveal private information, such as passwords, card numbers, etc. The plot of this attack is that the phishing website appears exactly the same as that of the legitimate one to avoid any kind of suspicion. However, the URL features of both websites are different. These differences can be a strong basis for classifying a phishing website accurately and effectively. In our research, we train these features on an Artificial Neural Network(ANN) for accurate classification. Pharming is a special type of phishing attack or DNS poisoning in which the user is redirected to a fake website by changing the IP address at the DNS server. We implement provisions to detect pharming to provide overall protection to the user.