Android has retained its global popularity among cell phone users. Simultaneously, there has been a growth of malware targeting the platform, with larger current lines using unexpectedly cutting-edge detection evasion strategies. Options for timely 0-day detection are necessary as standard signature-based approaches become less effective in detecting unknown threats. This study contributes a strategy based on ensemble learning for detecting malicious apps on Android. To improve the accuracy of Android malware detection, it combines the advantages of hybrid analysis with the efficiency and performance of ensemble device research. Machine learning models are constructed using a large collection of malicious and safe software. The built model can predict all real-world test scenarios. The user has complete access to all program information on their devices. The proposed system employs a variety of machine learning algorithms for ensemble learning, including Decision Tree, Random Forest, and Androguard. The testing module is intended to ensure the efficiency and efficacy of the proposed system. The results show that the suggested method is extremely effective at identifying Android malware, with a detection accuracy of up to 100%.