Malware Detection in Android Based Devices by a Hybrid Approach Using Machine Learning Techniques

Android applications are rapidly evolving across the mobile ecosystem, but they also face a growing threat from Android malware. Researchers have extensively analyzed the concept of detecting Android malware and proposed concepts and procedures from various perspectives. Machine learning has been identified as an effective and promising approach for detecting Android malware. However, some reviews have addressed various problems which are in relation with Android malware detection using machine learning. Therefore, our paper aims to provide a more comprehensive survey that covers a wider range of aspects of the topic. As part of this paper, we provide a comprehensive overview of Android malware detection approaches that rely on machine learning. We briefly introduce the background of Android apps, covering the Android system design, security procedures, and malware categorization. We analyze and summarize the research status from important angles such as sample collecting, data pre-processing, feature selection, machine learning models, algorithms, and detection efficacy, with a focus on machine learning. Finally, we examine the future possibilities for detecting malware in Android using Machine Learning. By covering an extensive variety of aspects, our review aims to help researchers gain a better understanding of Android malware detection using machine learning. It will serve as a foundation for future studies and help guide research in this field.