Jan, Salman, Ul Rehman, Shafiq, Syed, Toqeer Ali, ilani, Abdul Khadar, Khan, Muhammad Yasar and Ali, Jawad ORCID: https://orcid.org/0000-0002-6015-0663 (2025) Advancing Android Malware Detection: Innovative Approach in Behavioral Log Analysis. In: Hamdan, R. K., (ed.) Sustainable Data Management. Studies in Big Data (170). Springer, Cham, pp. 611-621

Full text not available from this repository.

Abstract

Securing a smartphone from malware intrusion is a critical issue that requires immediate attention. This study utilizes a Deep Convolutional Generative Adversarial Network (DCGAN) to introduce a novel approach to Android malware detection. The proposed model is trained on the behavioral logs of over 10,000 Android applications by focusing on critical resource consumption, which requires access to Android permissions such as SMS, camera, Internet, and location access. The proposed model successfully classified benign and malicious applications using discriminative features, achieving 96.5% accuracy, 97.0% precision, 96.1% recall, and a 96.5% F1-score. These evaluation metrics show that feature learning through DCGAN is better than traditional feature extraction techniques, which significantly reduce false positives. This study identifies the need for enhanced permission control and real-time malware monitoring and detection system in the existing Android security model. It proposes a scalable and adaptable solution to make android-malware threats, advancing the state-of-the-art in Android malware detection.

Item Type:	Book Section
Status:	Published
DOI:	10.1007/978-3-031-83915-3_50
Subjects:	Q Science > Q Science (General) > Q325 Machine learning
School/Department:	London Campus
URI:	https://ray.yorksj.ac.uk/id/eprint/14532

University Staff: Request a correction | RaY Editors: Update this record