CFSBFDroid: Android Malware Detection Using CFS + Best First Search-Based Feature Selection

With the fast development of smartphone technology and mobile applications, phone has become most powerful tool to access Internet get various services with one click. Meanwhile, susceptibilities application are primary hazard security Android devices. Due these weaknesses, an attacker can easily hack confidential data phone. The malware automatically performs fraudulent activities on phones without user's knowledge. Thus, attacks major threats phones. To detect malicious applications installed smartphones, we have conducted a study that focuses permissions intent-based mechanisms. was done in three phases: first phase, dataset created by extracting intents from APK files; second correlation-based feature selection (CFS) best search (BFS) were combined select representative features space extracted dataset; third machine learning (ML) techniques trained tested against preprocessed obtained phase. accuracy, precision, recall, F1 score, error metrics seven (REPTree, Rule PART, RF, SMO, SGD, MCC, LMT) demonstrated over dataset.