Android Malware Detection using Fast Correlation Based Filter and Random Forest Algorithm

Malware analysts have faced several difficulties in the past few years due to the rapid growth of Android malware. This surge in Android malware not only threatens the security and privacy of users but also undermines the integrity of the Android ecosystem. As cybercriminals get more advanced, the methods for detecting and dealing with malware need to grow more sophisticated. To find the best combination for identifying Android malware, this study compares different machine learning algorithms and feature selection techniques. The dataset used in this research is the CICMaldroid2020 dataset. The results indicate that the combination of the Fast Correlation-Based Filter feature selection method and the Random Forest algorithm achieved the highest performance, with both accuracy and F1-Score at 92.5%.