Sistem Deteksi Android Malware Menggunakan Metode Gated Reccurent Unit (GRU) Deep Learning

Abidin, Zainal (2023) Sistem Deteksi Android Malware Menggunakan Metode Gated Reccurent Unit (GRU) Deep Learning. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 05211940000102_Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (2MB) | Request a copy

Abstract

Malware merupakan perangkat lunak jahat yang dirancang dengan tujuan untuk merusak sistem device-device ( smartphone, computer, dan perangkat-perangkat lain ) korban. Penggunaan sistem operasi Android menempati urutan pertama dalam pangsa pasar. Android merupakan sistem operasi yang menggunakan kernel linux, linux merupakan sistem operasi yang open source. Sehingga dengan kondisi tersebut menyebabkan dengan lebih mudah dan lebih menguntungkan bagi pembuat malware untuk menargetkan operasi sistem yang opensource, dan pada android tidak adanya batasan pemasangan aplikasi dari sumber mana pun yang memungkinkan. Oleh karena itu diperlukan pendeteksi yang dapat mengikuti perkembangan serta menyadarkan pengguna kepeduliannya terhadap kemaanan informasi. Pada tugas akhir ini akan menggunakan dataset Android malware yang teridiri fitur permission, dataset tersebut meru pakan dataset opensource terdiri dari 2 kelas sampel ( malware dan benign) dengan total 11,975 sample (6,000 malware and 5,975 benign) dalam bentuk format file csv. Arsitektur model deep learning yang digunakan adalah arsitektur GRU .Pada penilitian tugas akhir ini akan membandingkan kinerja yang dihasilkan dari tiga optimizer yaitu adamax, radam, dan SGD. Kinerja yangdibandingkan meliputi precision , recall , f1-measure dan specitivity.
==================================================================================================================================
Malware is malicious software designed with the intent to damage the systems of devices such as smartphones, computers, and other equipment. Android operating systems hold the highest market share, making them a prevalent target for malware creators. As an open-source system using a Linux kernel, Android’s accessibility potentially provides more opportunities for malware developers to exploit. Moreover, Android allows applications to be installed from any source, adding to the vulnerability. Therefore, it is crucial to have detection systems that can evolve with the malware landscape and alert users to potential threats to their information security.In this research, we utilize an open-source Android malware dataset, which consist of permissions features. The dataset consists of two sample classes (malware and benign) with a total of 11,975 samples (6,000 malware and 5,975 benign), represented in a CSV file format. We employ a deep learning model architecture known as the Gated Recurrent Unit (GRU) .This research aims to compare the performance of three optimizers—adamax, radam, and Stochastic Gradient Descent (SGD)—in terms of precision, recall, F1-measure, and specificity. This com- parison aims to determine which optimizer produces the most efficient and accurate malware detection results for Android systems.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Android, Gated Recurrent Unit, Malware ,Optimizer.
Subjects:	T Technology > T Technology (General) > T174 Technological forecasting T Technology > T Technology (General) > T57.5 Data Processing T Technology > T Technology (General) > T57.8 Nonlinear programming. Support vector machine. Wavelets. Hidden Markov models. T Technology > T Technology (General) > T58.62 Decision support systems
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Information System > 57201-(S1) Undergraduate Thesis
Depositing User:	Zainal Abidin
Date Deposited:	30 Jul 2023 06:49
Last Modified:	30 Jul 2023 06:49
URI:	http://repository.its.ac.id/id/eprint/99166

Actions (login required)

View Item