MALWARE DETECTION USING DEEP LEARNING TO ANALYZE WITH DEEP NEURAL NETWORKS DEPLOYED ON THE CLOUD

Abstract

Malware attacks have become one of the most critical cybersecurity threats in modern digital environments due to the rapid growth of internet-connected systems, cloud computing, and smart devices. Traditional signature-based malware detection methods are increasingly ineffective against evolving threats such as zero-day attacks, polymorphic malware, and obfuscated malicious code. This study presents the design, implementation, and evaluation of a deep learning-based malware detection system integrated with cloud computing for scalable and real-time threat detection. The system leverages advanced artificial intelligence techniques, including Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM), and hybrid deep learning architectures, to automatically extract and learn patterns from malware datasets. The model was trained and tested using publicly available datasets such as EMBER and other cybersecurity repositories. Data preprocessing techniques, including normalization, feature extraction, and class balancing using SMOTE, were applied to enhance model performance. The proposed system was implemented using Python and machine learning frameworks such as TensorFlow and Keras, and deployed on a cloud platform to support real-time malware detection. Experimental results show that the system achieved high performance with an accuracy of 97.8%, precision of 96.9%, recall of 97.3%, and an F1-score of 97.1%, demonstrating its effectiveness in detecting both known and unknown malware threats. The findings confirm that deep learning-based approaches significantly outperform traditional methods in malware detection tasks. The integration of cloud computing further enhances scalability, efficiency, and accessibility. This study concludes that AI-driven malware detection systems provide a reliable solution for modern cybersecurity challenges and recommends further research into explainable AI, federated learning, and edge-based malware detection systems.