Survey on AI Malware Detection Methods and Cybersecurity Education

Abstract

This paper provides an extensive overview of recent advancements in deep learning-based methods for detecting malware and in programs for educating people about cybersecurity. The overview includes hybrid models, detection based on images, and advanced techniques for extracting features such as texts and images. The main techniques assessed include Convolutional Neural Networks (CNNs), Long ShortTerm Memory networks (LSTMs), and hybrid models that combine CNNs with Recurrent Neural Networks (RNNs). Furthermore, this article assesses strategies for cybersecurity education, with a focus on engaging employees, providing targeted education for at-risk groups, and integrating digital learning tools. While deep learning models have greatly enhanced the accuracy of malware detection, challenges like the quality of datasets, computational expenses, and adversarial attacks continue to exist. In the field of cybersecurity education, promoting awareness through interactive and gamified techniques has been proven to be effective in creating a more resilient workforce. This overview examines these challenges and suggests future directions, including hybrid models for improved malware detection and scalable digital tools for cybersecurity education