UB ScholarWorks

Proposed Framework to Improving Performance of Familial Classification in Android Malware

Show simple item record

dc.contributor.author Alswaina, Fahad A.O.
dc.date.accessioned 2020-10-27T13:45:34Z
dc.date.available 2020-10-27T13:45:34Z
dc.date.issued 2020-10-27
dc.identifier.citation F.A.O. Alswaina, "Proposed Framework to Improving Performance of Familial Classification in Android Malware", Ph.D. dissertation, Dept. of Engineering, Univ. of Bridgeport, Bridgeport, CT, 2020. en_US
dc.identifier.uri https://scholarworks.bridgeport.edu/xmlui/handle/123456789/4386
dc.description.abstract Because of the recent developments in hardware and software technologies for mobile phones, people depend on their smartphones more than ever before. Today, people conduct a variety of business, health, and financial transactions on their mobile devices. This trend has caused an influx of mobile applications that require users' sensitive information. As these applications increase so too have the number of malicious applications increased, which may compromise users' sensitive information. Between all smartphone, Android receives major attention from security practitioners and researchers due to the large number of malicious applications. For the past twelve years, Android malicious applications have been clustered into groups for better identification. Characterizing the malware families can improve the detection process and understand the malware patterns. However, in the research community, detecting new malware families is a challenge. In this research, a framework is proposed to improve the performance of familial classification in Android malware. The framework is named a Reverse Engineering Framework (RevEng). Within RevEng, applications' permissions were selected and then fed into machine learning algorithms. Through our research, we created a reduced set of permissions using Extremely Randomized Trees algorithm that achieved high accuracy and a shorter execution time. Furthermore, we conducted two approaches based on the extracted information. The first approach used a binary value representation of the permissions. The second approach used the features' importance. We represented each selected permission in latter approach by its weight value instead of its binary value in the former approach. We conducted a comparison between the results of our two approaches and other relevant works. Our approaches achieved better results in both accuracy and time performance with a reduced number of permissions. en_US
dc.language.iso en_US en_US
dc.subject Android malware en_US
dc.subject Artificial intelligence en_US
dc.subject Cybersecurity en_US
dc.subject Data science en_US
dc.subject Information security en_US
dc.subject Machine learning en_US
dc.title Proposed Framework to Improving Performance of Familial Classification in Android Malware en_US
dc.type Thesis en_US
dc.institute.department School of Engineering en_US
dc.institute.name University of Bridgeport en_US


Files in this item

This item appears in the following Collection(s)

Show simple item record

Search ScholarWorks


Advanced Search

Browse

My Account