Repositori Institusi

Politeknik Siber dan Sandi Negara

  • Home
  • Information
  • News
  • Help
  • Librarian
  • Member Area
  • Select Language :
    Arabic Bengali Brazilian Portuguese English Espanol German Indonesian Japanese Malay Persian Russian Thai Turkish Urdu

Search by :

ALL Author Subject ISBN/ISSN Advanced Search

Last search:

{{tmpObj[k].text}}
Image of Implementasi Deep Learning untuk Deteksi dan Klasifikasi Malware Android Menggunakan Analisis Berbasis Hybrid
Bookmark Share

Text

Implementasi Deep Learning untuk Deteksi dan Klasifikasi Malware Android Menggunakan Analisis Berbasis Hybrid

Girinoto - Personal Name; Ray Novita Yasa - Personal Name; R Budiarto Hadiprakoso - Personal Name; Arga Prayoga - Personal Name;

Abstrak:
Sistem operasi Android merupakan sistem operasi ponsel yang paling cepat berkembang karena sifatnya yang open-source. Namun, popularitas yang dimiliki Android tidak selalu berdampak positif bagi penggunanya. Salah satu ancaman bagi pengguna Android adalah adanya malicious software (malware). Oleh karena itu, tindakan preventif perlu dilakukan untuk mendeteksi dan klasifikasi malware. Salah satu cara yang dapat digunakan adalah deep learning. Penelitian ini membandingkan arsitektur single-view deep learning dan multi-view deep learning untuk mendeteksi dan klasifikasi Android malware dengan memanfaatkan analisis malware Android berbasis hybrid yaitu menggabungkan analisis statis dan dinamis untuk menutupi kelemahan dari keduanya. system-call sequence dan network flow diperoleh dari analisis dinamis sedangkan static feature diperoleh dari analisis statis dari sebuah aplikasi. Model deep learning yang dibangun menggunakan tiga algoritma berbeda yaitu Bi-LSTM dan Bi-GRU untuk memproses system-call serta MLP untuk memproses network flow dan static feature. Pada arsitektur single-view deep learning, masing-masing fitur diproses pada model secara terpisah, sedangkan pada multi-view deep learning ketiga fitur diproses pada sebuah model yang telah digabungkan dengan fungsi concatenate. Setelah melalui tahap training model dan evaluasi, hasil penelitian menunjukkan bahwa model terbaik terdapat untuk arsitektur binary classification adalah model Bi-GRU dan Bi-LSTM dengan hasil mencapai 100%. Sedangkan pada arsitektur multi-class classification, model MLP berdasarkan static feature memperoleh hasil sebesar 94%.
Abstract:
The Android operating system is the fastest-growing mobile operating system due to its open-source nature. However, the popularity of Android does not always have a positive impact on its users. One of the threats to Android users is the presence of malicious software (malware). Therefore, preventive measures need to be taken to detect and classify malware. One way to do this is through deep learning This research compares single-view deep learning and multi-view deep learning architectures for the detection and classification of Android malware by utilizing hybrid-based Android malware analysis, which combines static and dynamic analysis to cover the weaknesses of both. system-call sequence and network flow are obtained from dynamic analysis while static features are obtained from static analysis of an application. The deep learning models are built using three different algorithms, namely Bi-LSTM and Bi-GRU to process system-calls, and MLP to process network flows and static features. In the single-view deep learning architecture, each feature is processed separately in the model, while in multi-view deep learning, all three features are processed in a single model that has been combined with the concatenate function. After going through the model training and evaluation stages, the research results show that the best models for binary classification architecture are the Bi-GRU and Bi-LSTM models with a result of up to 100%. Meanwhile, in the multi-class classification architecture, the MLP model based on static features obtained a result of 94%.


Availability

No copy data

Detail Information
Series Title
--
Call Number
2023 ARG i
Publisher
Bogor : Politeknik Siber dan Sandi Negara., 2023
Collation
xviii, 91 halaman
Language
Indonesia
ISBN/ISSN
--
Classification
Rekayasa Perangkat Lunak Kriptografi
Content Type
-
Media Type
-
Carrier Type
-
Edition
--
Subject(s)
deep learning
Android Malware
Multi-View Deep Learning
Binary Classification
Multi-class Classification
Specific Detail Info
-
Statement of Responsibility
Arga Prayoga
Other version/related

No other version available

File Attachment
No Data
Comments

You must be logged in to post a comment

Repositori Institusi
  • Information
  • Services
  • Librarian
  • Member Area

About Us

Search

start it by typing one or more keywords for title, author or subject

Keep SLiMS Alive Want to Contribute?

© 2025 — Senayan Developer Community

Powered by SLiMS
Select the topic you are interested in
  • Computer Science, Information & General Works
  • Philosophy & Psychology
  • Religion
  • Social Sciences
  • Language
  • Pure Science
  • Applied Sciences
  • Art & Recreation
  • Literature
  • History & Geography
Icons made by Freepik from www.flaticon.com
Advanced Search
Where do you want to share?