DATDroid: Dynamic Analysis Technique in Android Malware Detection

摘要: Android system has become a target for malware developers due to its huge market globally in recent years. The emergence of 5G the and limited protocols post great challenge security Android. Hence, various techniques have been taken by researchers ensure high devices. There are three types analysis namely static, dynamic hybrid used detect analyze malicious application Due evolving nature malware, it is very challenging existing efficiently accurately. This paper proposed Dynamic Analysis Technique Malware detection called DATDroid. technique consists phases, which includes feature extraction, selection classification phases. A total five features call, errors time call process, CPU usage, memory network packets extracted. During 70% dataset was allocated training phase 30% testing using machine learning algorithm. Our experimental results achieved an overall accuracy 91.7% with lower false positive rates as compared benchmarked method. DATDroid also higher precision recall rate 93.1% 90.0%, respectively. Hence our proven be able classify more accurately reduce misclassification benign significantly.