Comparative Analysis of Machine Learning and Deep Learning Models for Android Malware Detection in Cybersecurity and Data Protection Contexts

The rapid growth of Android applications has led to an increased risk of malware attacks, posing significant threats to cybersecurity and personal data protection. This study proposes an artificial intelligence-based approach for Android malware detection using machine learning and deep learning techniques applied to permission-based features extracted from application data. A dataset consisting of Android application attributes was processed and used to train and evaluate Random Forest and Deep Neural Network models. The experimental results show that the deep learning model outperforms the Random Forest classifier, achieving an accuracy of 92.48 percent and a weighted F1-score of 0.9216, indicating its effectiveness in distinguishing between malicious and benign applications. In addition, feature importance analysis reveals that permissions related to system access, SMS handling, and background processes play a crucial role in identifying malware behavior. From a legal perspective, the findings highlight the potential of AI-driven malware detection systems to support cybersecurity compliance and data protection regulations by enabling early threat identification. However, challenges such as class imbalance, potential misclassification, and the need for transparent decision-making emphasize the importance of integrating technical solutions with legal safeguards. Overall, this study demonstrates that AI-based approaches can enhance Android malware detection while contributing to the enforcement of cybersecurity and data protection principles.