Malware steals the privacy-sensitive data stored in mobile devices and presents a serious threat to the security of devices and the services provided. Detection and analysis of Android malware have become one of the fastest-growing technology fields in recent years. In particular, Android-based IoT devices have become one of the major targets of malware attacks due to the openness and universality of the Android platform.
Existing malware detection methods heavily rely on the accumulation of signature libraries and human intervention by malware analysts; thus, it has been a challenging task to adapt to the explosive growth of Android malware. Deep learning methods for android malware detection detect malware without manual feature engineering. Thus, Deep Ensemble learning is an effective way for Android malware detection. It employs individual accuracy and diversity within the ensemble simultaneously and raises a fusion model to learn the implicit information from the output of ensemble learners to optimize the classification performance further.