Phishing detection using Machine learning and Deep learning techniques: A review

Abstract

Cyber criminals have not ceased to develop new ways of attacking users of the cyberspace, this is sees in the continuous style of phishing attacks. This paper aims at providing a comprehensive overview of the latest state-of-art in the development and implementation of phishing detection system with particular emphasis on the use of machine learning and deep learning methods. The review includes a wide variety of studies, comparing the characteristics of various system architectures, algorithms, applications, and implementation tools used in the development of recommender systems. Discovery includes massive trends in development of new combined deep learning models, an identification of feature engineering and data preprocessing as significant steps in developing DL models, and identification of brand-new approaches to ease the learning of deep learning such as visual similarity analysis, and semantic analysis of texts. The review also notes that heightened user awareness and education helps make users who are targeted by phishing more resilient while pointing out that the benefits of blockchain technology in improving cyberspace security and meaningfully advancing cooperation continue to remain enormous. This paper offers an extensive assessment of the existing literature on phishing detection, as well as sheds light on possible future research directions that are more effective in fight against this increasingly looming menace.