Thesis: |
Android Malware Classification Based on Mobile Security Framework
|
Supervisor: | Assistant Professor Worawat Choensawat (Dr.) and Dr. Romuald Jolivot |
Graduate Year: | 2017 |
Abstract: |
A large number of smartphones are running android operating system. Android is an open source platform and hence has become very popular. This high popularity is also attracting many rouge authors. In this study, a machine learning based technique is proposed to classify android applications in three classes based on the confidence level defined as safe, suspicious and highly suspicious. The confidence level refers to how likely an application is a malware. For the development of the classification model 36 features are extracted and selected using Mobile Security Framework based on penetration testing. A set of experiments has been conducted using 14,073 android applications which consists of 10,000 android applications automatically downloaded from apk-dl.com android market place, 3041 known malware and 1032 benign applications. In order to compare the accuracy of the classification model, a benchmark is created using VirusTotal. The proposed method can detect and classify android applications into three confidence levels with 81.8% accuracy. However, to identify the effectiveness of selected features one more experiment is conducted based on binary (detect an application of being malware or benign) classification which has yielded 93.63% accuracy. |
Keywords : | cybersecurity, android operating system, anti-virus, mobile operating systems, pattern recognition |
Current Position: | - |
Institute/Company: | - |
email: | This email address is being protected from spambots. You need JavaScript enabled to view it. |
Web: | - |
Independent Study: | Online Animation Displaying from Motion Capture Database |
Supervisor: | Assistant Professor Worawat Choensawat (Dr.) and Dr. Romuald Jolivot |
Graduate Year: | 2017 |
Abstract: |
A large number of smartphones are running android operating system. Android is an open source platform and hence has become very popular. This high popularity is also attracting many rouge authors. In this study, a machine learning based technique is proposed to classify android applications in three classes based on the confidence level defined as safe, suspicious and highly suspicious. The confidence level refers to how likely an application is a malware. For the development of the classification model 36 features are extracted and selected using Mobile Security Framework based on penetration testing. A set of experiments has been conducted using 14,073 android applications which consists of 10,000 android applications automatically downloaded from apk-dl.com android market place, 3041 known malware and 1032 benign applications. In order to compare the accuracy of the classification model, a benchmark is created using VirusTotal. The proposed method can detect and classify android applications into three confidence levels with 81.8% accuracy. However, to identify the effectiveness of selected features one more experiment is conducted based on binary (detect an application of being malware or benign) classification which has yielded 93.63% accuracy. |
Keywords : | cybersecurity, android operating system, anti-virus, mobile operating systems, pattern recognition |
Current Position: | - |
Institute/Company: | - |
email: | This email address is being protected from spambots. You need JavaScript enabled to view it. |
Web: | - |