Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/74923
Title: Deep learning of large-scale android malware detection
Authors: Loo, Jia Yi
Keywords: DRNTU::Engineering
Issue Date: 2018
Abstract: Smartphones had brought convenience to our lives. However, malware attacks can easily disrupt this convenience. Given the large Android market size and its vulnerability to malware, this report will focus on Android malware detection by the means of machine learning and deep learning. In this report, a re-implementation of a newly proposed machine learning method was done and tested on large real world datasets. Extensive simulations had been conducted with various parameters. The simulations included classification of applications into benign and malware, using both machine learning and deep learning methods, the clustering of malware families and clone applications. It was found that classification using machine learning was more efficient and accurate than that of deep learning. As for the two clustering applications, after a series of experiment, it was concluded that Agglomerative clustering model with ward linkages was the best model to be used. The findings obtained would give a more detailed understanding of the behaviour of malware applications as well as the types of methods suitable for Android malware detection. Due to some limitations, it is recommended that more simulations to be done so as to give even more detailed findings.
URI: http://hdl.handle.net/10356/74923
Rights: Nanyang Technological University
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:EEE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
Final Report.pdf
  Restricted Access
FYP Final Report1.94 MBAdobe PDFView/Open

Page view(s) 20

282
checked on Oct 24, 2020

Download(s) 20

37
checked on Oct 24, 2020

Google ScholarTM

Check

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.