AFWDroid: Android Kötücül Yazılım Tespitinde Derin Özellik Çıkarımı ve Ağırlıklandırma

Recep Sinan Arslan,Orhan Er,Emre Ölmez

doi:10.24012/dumf.875036

Abstract

Android malware detection is a critical and important problem that must be solved for a widely used operating system. Conventional machine learning techniques first extract some features from applications, then create classifiers to distinguish between malicious and benign applications. Most of the studies available today ignore the weighting of the obtained features. To overcome this problem, this study proposes a new software detection method based on weighting the data in feature vectors to be used in classification. To this end, firstly, the manifest file was read from the Android application package. Different features such as activities, services, permissions were extracted from the file, and for classification, a selection was made among these features. The parameters obtained as a result of selection were optimized by the deep neural network model. Studies revealed that through feature selection and weighting, better performance values could be achieved and more competitive results could be obtained in weight-sensitive classification.

Highlights

Development of mobile internet technologies, the prevalence of mobile devices is gradually increasing, and Android is leading this increase [1]
In the first quarter of 2020, it was determined that an average of 480.000 malware appeared per month, showing a significant increase [5]. This structure offered by the Android system on the hardware and software side as well as the huge user base mentioned above has caused the emergence of malicious application developers for this operating system, the development of applications to exploit end-users with little experience of usage and increase in efforts to obtain information illegally
We propose a hybrid approach using static analysis and machine learning techniques for malware detection

Summary

Introduction

Development of mobile internet technologies, the prevalence of mobile devices is gradually increasing, and Android is leading this increase [1]. In the first quarter of 2020, it was determined that an average of 480.000 malware appeared per month, showing a significant increase [5] This structure offered by the Android system on the hardware and software side as well as the huge user base mentioned above has caused the emergence of malicious application developers for this operating system, the development of applications to exploit end-users with little experience of usage and increase in efforts to obtain information illegally. In the Android operating system, software developed with different ROMs and applications distributed from third parties are stored on end-user mobile devices This requires the development of systems that use the most up-to-date methodologies to restrict or prevent access to sensitive personal information, to detect malware, and to secure mobile devices. The study has been evaluated in general and suggestions for new studies were given

Related Works

Methodology

Evaluation Criteria

Conclusion and Future Works