Abstract: Code analysis and Malwares detection for Android applications are considered as a serious problem. There are many researches to apply new and creative techniques that can detect Malwares with low computational cost. These researches are being grown because of wide use of new applications. This thesis tries to attack this problem by presenting a new and creative method to analyze static code and measure similarity with available dataset of applications. Thesis immigrate its idea to a well-known and wide officially used antivirus project, which is considered as a part of famous antiviruses programs and called Androguard. We apply thesis idea in Androguard project to test its feasibility by making a comparison study between it and modified Androguard; we used Malware dataset of android applications for this research. As a result, proposed method saves about 60-70% of time with similar results and time distribution behavior in compare with original Androguard and little scarification in accuracy, this refer to simplicity of generating signatures and measuring similarity using SimHash algorithm.
Keywords: Androguard, similarity, applications, android applications, researches.
Title: Similarity among Android Applications by GUI Feature Extraction
Author: Amit Kumar, Vishal Verma
International Journal of Computer Science and Information Technology Research
ISSN 2348-1196 (print), ISSN 2348-120X (online)
Research Publish Journals
