Comparative Analysis of Deep Learning Models for Predictive Phishing Attacks and Virus Message Forecasting
Open Access DepositedPhishing attacks have become more sophisticated and successful in recent years due to the advanced social engineering techniques, curated emails that look legitimate, and the manipulative skills attackers have on their target. This has become a threat to government organizations due to their hold on confidential data. Government employees have become a desired target to access this sensitive information (Alkhalil et al., 2021). Traditional machine learning models can effectively predict the likelihood of a phishing attack and be able to detect emails containing malicious content however, deep learning techniques especially when combined to form a hybrid model may yield more accurate results than the traditional supervised method. Utilizing deep learning techniques and combining them for a hybrid approach is a new method that can be adopted by organizations. To find the most desired model is still being studied. In order to gain proper results, it is important to incorporate the appropriate model based on the data that is being used. Deep learning models offer promising capabilities to forecast the likelihood of a phishing attack. These models continue to adapt and evolve which results in better prediction and detection results, helping mitigate an impact of a phishing attack. This study demonstrates that deep learning models, particularly when combined with a hybrid model, can produce better outcomes when predicting future failures in a phishing resilience assessment exercise and how deep learning forecasting models can efficiently predict the number of viral messages.
- All rights reserved
Notice to Authors
If you are the author of this work and you have any questions about the information on this page, please use the Contact form to get in touch with us.