An efficient approach for detection and analysis of Android malware via machine learning

An efficient approach for detection and analysis of Android malware via machine learning


An efficient approach for detection and analysis of Android malware via machine learning

نوع: Type: thesis

مقطع: Segment: masters

عنوان: Title: An efficient approach for detection and analysis of Android malware via machine learning

ارائه دهنده: Provider: arezoo yavari yeganegi

اساتید راهنما: Supervisors: mahdi sakhaei nia

اساتید مشاور: Advisory Professors: moharram mansoori zahdeh

اساتید ممتحن یا داور: Examining professors or referees: yousef sanati, reza mohammadi

زمان و تاریخ ارائه: Time and date of presentation: 2023

مکان ارائه: Place of presentation: class 27

چکیده: Abstract: With the increase in the use of mobile devices, malware attacks increase, especially in Android phones, which occupy a large share of the market. Hackers try to attack smartphones with different methods such as credit theft, surveillance and malicious advertisements. There are also malicious programs that are expanding and they are called malware that can steal important user information, make calls and send messages without the user's permission. Therefore, many solutions have been created to discover these malwares by using static analysis of program codes and checking the permissions given, or dynamic analysis and program execution and checking the behavior of the program, or a combination of these two methods. The ability to identify malware and examine the main characteristics of programs. Among the many countermeasures, machine learning-based methods have been recognized as an effective means of detecting these attacks because they can classify a set of training examples. , so they do not need to explicitly define the signature. In this thesis, from a data set with 5560 malwares, 100 malwares are randomly selected 10 times and each time 100 malwares are collected from each of these sets along with 2201 healthy applications and their characteristics. Then these features are given to machine learning algorithms. To distinguish between malware and healthy applications

فایل: ّFile: