A Static Feature Selection-based Android Malware Detection Using Machine Learning Techniques
Aviral Sangal, Harsh Kumar Verma
Abstract
With an increase in popularity and usage of smartphones, attackers are constantly trying to get sensitive information from smartphones. To protect the information, researchers are constantly working on the effective detection of android malware. Since there has been a large-scale increase in the number of new malware being detected, machine learning based techniques have to turn towards for effective large-scale detection. In this paper, CICInvesAndMal2019 have been taken as dataset and used android permissions and intent as a feature set for malware detection. Principal Component Analysis was used as a feature selection approach The dataset is trained and tested over well-known machine learning models and Random Forest was the best classifier with 96.05% accuracy.