Machine Learning for Malware Detection on Android: Systematic Review of the Literature

Cárdenas-Quispe A.; Vergaray-Mezarina R.; Gamboa-Cruzado J. · RISTI - Revista Iberica de Sistemas e Tecnologias de Informacao (2021)

Resumen

Android has become the most attractive target for malware attacks. This paper presents a systematic literature review on the research trends in Android malware detection based on Machine Learning. The results of the systematic review have concluded prominent trends of Machine Learning algorithms, criteria for measuring effectiveness and other operating systems where Machine Learning is applied for malware detection. Statistical results show that Support Vector Machine (23.14%), Random Forest Regression (16.16%), Naive Bayes Classifier (14.85%), J48 Decision Tree (10.92%) and k-NN (10.48%) algorithms are the most used in Android malware detection. This indicates that, the increasing popularity and massification of Android implies the research of cutting-edge detection techniques based on Machine Learning for a more effective detection.

Instituciones: Universidad Nacional Federico Villarreal