Intelligent Systems for Malware Detection on Android Devices: A Comparative Analysis of Static and Dynamic Analysis Approaches

Abstract

This article explores intelligent systems for detecting malware on Android devices, focusing on static and dynamic analysis techniques. It provides a detailed comparison of these approaches in terms of accuracy, performance, resource usage, real-time capabilities, and resistance to obfuscation. Static analysis examines an app’s code and permissions without execution, while dynamic analysis observes its runtime behavior in a controlled environment. The paper highlights how machine learning and deep learning models enhance detection accuracy for both methods and discusses their integration into modern malware detection systems. A comparative table and conceptual figures illustrate the trade-offs between static and dynamic methods. The article concludes by addressing key challenges such as evasion, resource limitations, and adversarial attacks, and outlines future research directions including hybrid analysis and explainable AI. This work serves as a reference for educators and researchers in the fields of mobile security and artificial intelligence.