Politeknik Siber dan Sandi Negara

Knowledge Center of Cybersecurity and Cryptography

  • 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 EfficientNet pada klasifikasi malware BIG2015
Bookmark Share

Text

Implementasi EfficientNet pada klasifikasi malware BIG2015

Handhika Yanuar Pratama - Personal Name; Jeckson Sidabutar - Personal Name; I Komang Setia Buana - Personal Name; Tiyas Yulita - Personal Name;

Malware merupakan program yang berpotensi merusak, menganggu kinerja, maupun melakukan pencurian data dari suatu sistem. Tercatat terjadi peningkatan laju persebaran malware sebesar 55% dalam lima tahun terakhir. Salah satu upaya untuk mengatasi persebaran ini yaitu dengan menggunakan antivirus. Dalam perkembangannya antivirus menggunakan teknik signature-based dan heuristic-based untuk pendeteksian dan klasifikasi malware. Tetapi kedua teknik tersebut sudah tidak dianggap relevan salah satunya semenjak kemunculan malware heuristic yang dapat mengubah strukturnya setiap dijalankan. Oleh karena itu kedua pendekatan tersebut ditinggalkan dan digantikan dengan pendekatan machine learning dan deep learning yang terbukti memiliki kecepatan dan akurasi yang lebih baik dalam klasifikasi malware. Penelitian ini melakukan klasifikasi malware dengan sampel malware BIG 2015 yang divisualisasikan menggunakan teknik B2IMG yang telah digunakan pada penelitian sebelumnya. Hasil visualisasi diklasifikasikan menggunakan model transfer learning yaitu EfficientNet. Model EfficientNet terbukti dapat mengalahkan state-of-the-art pada klasifikasi ImageNet. Hasil penelitian menunjukkan bahwa implementasi model ini memiliki performa lebih baik dibandingkan dengan beberapa teknik yang telah diimplementasikan sebelumnya seperti LSTM, Xception, dan RNN dengan memperoleh nilai accuracy sebesar 99.63%, precision sebesar 98.36%, recall 98.35%, F1-Score 98.34%, dan AUC 98.30%, dimana proses pembuatan model ini hanya menggunakan epochs sebanyak 10 kali. Hasil ini dapat digunakan sebagai pertimbangan dalam klasifikasi malware pada penelitian selanjutnya. --

Malware is a program that has the potential to damage, disrupts performance, or steal data from a system. There has been an increase in the growth rate of malware spreading by 55% in the last five years. One of the most used efforts to overcome the spread of malware is using an antivirus. In its development, antivirus uses signature-based and heuristic-based techniques for malware detection and classification. However, these two techniques are no longer considered relevant. One of the reasons is the growth of heuristic malware that can change its structure every time it is run. Therefore both approaches are being abandoned and replaced by machine learning and deep learning approaches that proved to have better speed and accuracy for malware classification. This study performs malware classification with BIG 2015 malware samples visualized using the B2IMG technique used in previous studies. The visualization results are classified using the transfer learning model, namely EfficientNet. EfficientNet is proven to beat the state-of-the-art classification of ImageNet. The results showed that this model was better than several previously implemented techniques such as LSTM, Xception, and RNN with an accuracy value of 99.63%, precision of 98.36%, recall of 98.35%, F1-Score 98.34%, and AUC 98.30%, where the process of making this model only uses epochs ten times. These results can be considered when classifying malware in future studies.


Availability
#
Rekayasa Keamanan Siber 2022 HAN i
TA20220101520
Available - Read on Location
#
Rekayasa Keamanan Siber 2022 HAN i
TA20220101521
Available - Read on Location
Detail Information
Series Title
-
Call Number
2022 HAN i
Publisher
Bogor : Politeknik Siber dan Sandi Negara., 2022
Collation
xvi, 125 hlm.
Language
Indonesia
ISBN/ISSN
--
Classification
--
Content Type
-
Media Type
-
Carrier Type
-
Edition
--
Subject(s)
Malware
deep learning
image processing
EfficientNet
Specific Detail Info
-
Statement of Responsibility
Handhika Yanuar Pratama
Other version/related

No other version available

File Attachment
No Data
Comments

You must be logged in to post a comment

Politeknik Siber dan Sandi Negara
  • Information
  • Services
  • Librarian
  • Member Area

About Us

Perpustakaan Politeknik Siber dan Sandi Negara menyediakan berbagai macam koleksi seperti Buku, Jurnal, Majalah, Koran, Referensi dan Konten Lokal.

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?