An Ensemble Approach Based on Fuzzy Logic Using Machine Learning Classifiers for Android Malware Detection

In this study, a fuzzy logic-based dynamic ensemble (FL-BDE) model was proposed to detect malware exposed the Android operating system. The FL-BDE contains structure that combines both processing power of machine learning (ML)-based methods and decision-making Mamdani-type inference system (FIS). structure, six different methods, namely, logistic regression (LR), Bayes point (BPM), boosted decision tree (BDT), neural network (NN), forest (DF) support vector (SVM) were used as ML-based benefit from their scores. However, through an approach involving process voting routing, scores only three which more successful in classifying either negative instances or positive sent FIS be combined. During combining process, processed incoming inputs determined malicious application score. Experimental studies performed by applying balanced dataset obtained APK files downloaded Drebin database Google Play Store. results showed us had much better performance than models did, with accuracy 0.9933, recall 1.00, specificity 0.9867, precision 0.9868, F-measure 0.9934. These also proved can competitive powerful detection compared those similar literature.