PERFORMANCE ENHANCEMENT IN ANDROID MALWARE DETECTION USING DIMENTIONALITY REDUCTION

Abstract

Android is one of the most common operating systems in smart phones in current scenario. People are investing in smart phones and which in turn is leading to an exponential increase in android users worldwide. As the number of android users are increasing day by day, there is a wide range of target audience for malicious applications and related issues. By using Android Malware detection, I can detect such malicious applications and prevent the users from becoming the target of such malicious applications. There are various strategies that have been already proposed for detecting Android malware have been proposed in the literature, there is still a need for attribute selection methods to be employed in Android malware detection systems. A machine learning-based malware detection method is suggested in this work to differentiate Android malware from benign apps. As the dimensionality reduction stage is introduced in this study to reduce the dimensions in a dataset and to decrease the time of training phase. When the results are obtained, it is observed that the accuracy of the mode is also improved by using dimensionality reduction.