Abstract
Systems and Networks can be seriously threatened by malware, often known as malicious software. The sophistication of malware assaults is increasing, making it harder to identify and stop them, for several reasons, including the protection of private data, data loss and alteration, system interruptions, monetary losses, and reputational harm. Therefore, malware detection and prevention are essential. Various machine learning models such as Random Forest, Support Vector Machine, K-NN, Extra Tree classifier, Gradient Boosting, and AdaBoost are applied for Android malware detection, as presented in this research. A Python-based machine learning tool called Python Optimised ML Pipeline (TPOT) uses genetic programming to maximize network throughput. To retrieve static information like permissions, network calls, API calls, and system traffic from the malicious apps for Android dataset, we employ TPOT to construct models. Moreover, a comparison has been made with traditional Machine learning classifiers and Automated ML for greater performance by reduction in computational time, training speed, and efficiency. Subsequently, metrics such as precision, F1-score, Recall, and accuracy are used to evaluate the overall performance of models. The analysis proved that Automated ML provides better outcomes of 99.7% accuracy with lesser computational complexity and a reduction in training time.