Enhanced App Model for the Prevention and Prediction of Malware in Android System

Malware is termed as malicious software, its aim is to cause harm to the mobile devices, server and also computer network by either collapsing the system or leakage of some confidential data. It does the damage when it is some way implanted into the mobile. The effects of the malware is that your mobile becomes unstable, it will continuously get reboot or it may also get crash. To overcome this problem an enhanced app model is designed. This app is efficient enough to detect all sort of malware and prevent the cell phones from getting security breached. To remove this malware, an antivirus is required which has definition of all the malware types but in this app we have used a Droid Fusion framework which is used to classify the malware types and their versions. It also helps to improve the accuracy to predict the malware in the devices and also predicts at what limit it will do harm to the mobile. Data flow API features are extracted to check whether the sensible data leaves out the system. In the previous system the implementation is done using the k-nearest algorithm but in this system we have used an enhanced Quine mccluskey algorithm technique to track and differentiate the malicious data from the normal data. Because k map based technique breakdown in six variables, Quine mccluskey proposed an algorithmic based technique for simplifying Boolean logic functions and basically it has two advantages over the kmap method. Firstly it is systematic for producing a minimal function that is less dependent on visual pattern it is viable scheme for handling large no .of variables.