Enhancing Malware Detection Using Deep Learning: A Study on MalNet Plus Architecture
Md Mojahidul Islam, Avishek Roy, Abir Hossain Apon, Prantha Chandra Sarkar, Vinoth Kumar R · 2024
The COVID-19 pandemic has accelerated the digital transformation of human life, leading to an increased presence of virtual environments. However, this shift has also attracted the attention of cybercriminals who exploit online platforms for criminal activities. Malicious programs, commonly known as malware, are employed by cyber attackers to carry out effective and evasive cyber-attacks. Traditional machine learning (ML) methods are struggling to cope with the complexities of advanced malware obfuscation techniques, rendering them less effective in malware detection. In this paper, we propose an enhanced convolutional neural network (CNN) architecture, named “MalNet Plus,” for malware classification. The MalNet Plus model is designed to improve feature extraction, feature optimization, and classification performance. It incorporates several key enhancements, including increased model depth, batch normalization, larger filter sizes, and normalization methods. Experiments conducted using the Malimg malware datasets demonstrate the superiority of the MalNet Plus model over other deep learning-based malware detection techniques. The model achieves an impressive accuracy rate of 98%, showcasing its effectiveness in accurately identifying and classifying malware samples. These findings emphasize the potential of deep learning techniques, specifically the MalNet Plus CNN architecture, in significantly enhancing malware detection accuracy and mitigating the rising cyber threats.