Cite Article

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

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

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