Ensemble of Artificial Neural Network and Deep Learning for the Prevention of Real-Time Phishing of Financial Data: A Systematic Literature Review

Source abstract

This study systematically reviewed relevant literatures on unsupervised machine learning models that have been used for detecting phishing in real-time. Phishing has caused loss of funds, exposed victims to future attacks and caused loss of revenue for financial product users as well as the financial industry at large. Various methods such as Self-Organizing Map, Generative Adversarial Network, Autoencoder, GMM, DBSCAN among many others were reviewed. Among the various methods for preventing phishing in real-time, ensemble model of deep learning and artificial neural network worked better and is the future of anomaly detection that researchers can even use in novel approach, hence it is suggested that future studies should apply the ensemble model used in this study for spamming. A good way of reducing phishing attack is for organizations to train staff who are users of devices, network and systems on how to detect phishing and defend against it. The players in the financial sector needs to understand that trend of technology has changed, hence, there is need for investment in phishing prevention technology and all other forms of cybercrimes.