MalDet: Malware Detector in Mobile Phone using Multilayer Perceptron Integrated with Virus Total

Abstract

The proliferation of Android applications has led to an increase in malicious software targeting mobile devices especially Android, posing significant security threats to users. This study presents a comprehensive approach to malware detector application in mobile applications using machine learning approach which is multilayer perceptron and virus total integration. By analysing static features extracted from various Android applications, the proposed system classifies applications as benign (safe) or malware (malicious). The methodology incorporates static analysis to extract critical features from the application package files (APKs). The feature focuses on permissions. Machine learning algorithms, Multilayer Perceptron, are employed for feature classification. Datasets chosen which is the combination of malware and benign apps are utilized to train and test the models, ensuring a robust evaluation of their performance. The integration of feature selection and optimization techniques further enhances detection accuracy, with results demonstrating high efficacy in identifying malicious applications. The integration with Virus Total providing an additional layer of verification for each scanned application. This research underscores the potential of machine learning in bolstering mobile security by providing an efficient and scalable solution for detecting Android malware.