Framework for the Detection of Android Malware using Artificial Immune System

Abstract

Artificial immune systems (AIS) are just computational systems that are inspired by theoretical immunology, observed immune functions, principles and mechanisms to solve problems including the detection of malware. AIS was used as optimizer for the selection of best features of android application. The aim of this paper is to propose an android malware classification technique for the detection of android malicious applications. The proposed framework consists of the basic approach and techniques to achieve good model for the detection of android malicious applications. The research methodology of Data Analysis, which involves validation through experimentation, is employed to achieve this. The results show that the models of selected permission-based features are more accurate than those models without the selection of features. The true positive rate and false alarm rate of selected features are also in better forms than those of classifying features without selection