Add to Favourites
To login click here
Researchers are putting efforts into providing reliable and resilient solutions for automatically detecting phishing attacks. Current ML-based phishing detection techniques are classified based on features used for detection, such as URL features, content features, and visual and hybrid features based detection. These techniques are discussed in the following section, with deep learning-based methods such as convolutional neural networks (CNN) and generative adversarial networks (GAN) being used to solve the data bias problem caused by the imbalance in phishing datasets. Sherazi et al. observed that URL is not the best way to detect a phishing website, and proposed a system that uses only domain names.