Detecting phishing attacks using Convolutional Neural Network and LSTM

Cybersecurity threats are increasing immensely due to the evolution of digitization. Due to this evolution, many security threats are happening around the world. One serious threat is phishing. It is a fraudulent activity that aims to steal a victim's private data using fraudulent emails and websites. Sensitive data such as login information, passwords, and bank account details are obtained and misused. In these types of attacks, the attackers appear to be legitimate. It is very hard to differentiate between what is real and phishing. A lot of research has been carried out to classify URLs as legitimate or related to phishing. However, a more efficient detection approach is needed to accurately predict and distinguish genuine sites from phishing sites. This issue can be solved by using deep learning algorithms such as convolutional neural networks (CNN) and long-term short-term memory (LSTM). This paper proposes a hybrid (CNN-LSTM) approach that provides maximum accuracy.