Unveiling the Efficacy of BERT’s Attention in Memory Obfuscated Malware Detection

Md Mashrur Arifin, Troy Suyehara Tolman, Jyh Haw Yeh

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

The study addresses the challenge of detecting obfuscated malware, particularly memory-obfuscated variants, which evade conventional detection methods by targeting a system’s volatile memory. By leveraging transformer-based models, notably BERT, the research demonstrates promising advancements in malware detection. Through data augmentation and rigorous feature selection processes, the study enhances the CIC-MlMem-2022 dataset, improving its quality for training classification models. Comparative analysis with conventional machine learning techniques highlights the superior performance of BERT and DistilBERT, achieving approximately 74% accuracy in classifying malware families. Notably, BERT exhibits exceptional capability in generalizing to unseen malware, achieving a remarkable 100% success rate in categorizing new families. These findings underscore the potential of transformer-based models for effectively detecting obfuscated malware, emphasizing the need for intelligent detection mechanisms in cybersecurity.

Original languageEnglish
Title of host publicationInformation Security Practice and Experience - 19th International Conference, ISPEC 2024, Proceedings
EditorsZhe Xia, Jiageng Chen
PublisherSpringer Science and Business Media Deutschland GmbH
Pages273-291
Number of pages19
ISBN (Print)9789819790524
DOIs
StatePublished - 2025
Event19th International Conference on Information Security Practice and Experience, ISPEC 2024 - Wuhan, China
Duration: 25 Oct 202427 Oct 2024

Publication series

NameLecture Notes in Computer Science
Volume15053 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference19th International Conference on Information Security Practice and Experience, ISPEC 2024
Country/TerritoryChina
CityWuhan
Period25/10/2427/10/24

Keywords

  • BERT-based Models
  • Malware Detection
  • Memory Obfuscated Malware

Fingerprint

Dive into the research topics of 'Unveiling the Efficacy of BERT’s Attention in Memory Obfuscated Malware Detection'. Together they form a unique fingerprint.

Cite this