Home About Login Current Archives Announcements Editorial Board
Submit Now For Authors Call for Submissions Statistics Contact
Home > Archives > Volume 20, No 11 (2022) > Article

DOI: 10.14704/nq.2022.20.11.NQ66049

An Effectual Analytics and Approach for Avoidance of Malware in Android using Deep Neural Networks

Kapil Aggarwal, Santosh Kumar Yadav

Abstract

Due to the rise in smartphone apps and Android use by people who use their phones a lot, there are a lot of security issues. Security issues need to be addressed in order to prevent vulnerabilities and find them before they happen. People who use smartphones are linked to a warning about the risks. Most people who use mobile phones don't have to think about a few bad things when they install APK files from different sources. It is important to make and use a system that can tell if code in Android apps is bad. Our first step is to look at the Android APK datasets. Both good and bad APKs are analyzed and dataset is processed. In this study, we'll look for and extract the signatures that are hidden in the APKs. This will make it easier to build a training dataset. As a whole, we're going to look at assorted dataset of APK files, and we think that about half of them will be safe and the other half will be dangerous. Then, we check to see if each APK has the permissions it needs and how it affects the way it works. Once it's been cleaned, a dataset will be made so that the model can be trained so that it can predict what will happen. To finish predictive analytics, any APK outside of the APKs is chosen and used. That's when it's possible to figure out how likely it is that the new APK being looked at will have bad code in it. Machine learning is used to track the results of different prediction measures, such as how long it takes, how accurate they are, and how much they cost. To compare two things, we use machine learning to combine our predictions. This article describes a machine learning technique to solve functional selection by safeguarding the selection and mutation operators of genetic algorithms. During population calculations in the training set, the proposed method is adaptable. Furthermore, for various population sizes, the proposed method gives the best possible probability of resolving function selection difficulties during training process. Furthermore, the proposed work is combined with a better classifier in order to detect the different malware categories. The proposed approach is compared and validated with current techniques by using different datasets.Using this approach, the accuracy is compared and found the elevated results in proposed approach.

Keywords

Android malware, apk analytics, android apk fingerprinting, smartphone security.

Full Text

PDF

References

?>