Classification and Detection of Malware in Android: An Analysis
Chander Prabha, Abhishek Mittal, Aarul Juneja · 2023
In the 21stcentury, the usage of the android operating system has increased at a very high rate thus thereby an increase in the number of subsequent apps that span and are used in android. As the number of android devices are increasing, the number of malicious actors producing android malware applications, which easily get downloaded on devices, is also increasing which makes the fight between security analysts and malware developers never-ending. Due to an increase in the number of apps, it is necessary to classify the apps either as benign or malware affected. The variation of android malware applications is being tackled by using machine learning and deep learning. This paper presents the detection and prediction of malware applications based on their app permission using machine learning algorithms. Further, their performance in classifying the apps is analyzed in terms of accuracy, recall score, and ROC scores. Results indicate that XG Boost and multilayer perceptron have higher accuracy of 98%. However random forest has a higher recall score of 99%.