• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to previous page

Combining deep learning and virtualization technologies to defend against banking malware

Research Project

Project/Area Number 20K21788
Research Category

Grant-in-Aid for Challenging Research (Exploratory)

Allocation TypeMulti-year Fund
Review Section Medium-sized Section 60:Information science, computer engineering, and related fields
Research InstitutionThe University of Tokyo

Principal Investigator

Shinagawa Takahiro  東京大学, 情報基盤センター, 准教授 (40361745)

Project Period (FY) 2020-07-30 – 2022-03-31
Project Status Completed (Fiscal Year 2021)
Budget Amount *help
¥6,500,000 (Direct Cost: ¥5,000,000、Indirect Cost: ¥1,500,000)
Fiscal Year 2021: ¥3,120,000 (Direct Cost: ¥2,400,000、Indirect Cost: ¥720,000)
Fiscal Year 2020: ¥3,380,000 (Direct Cost: ¥2,600,000、Indirect Cost: ¥780,000)
Keywords仮想化技術 / セキュリティ / 深層学習 / 機械学習 / マルウェア
Outline of Research at the Start

本研究では、バンキングマルウェアの特性に着目して、高精度な学習データが生成可能でかつ仮想化ソフトウェアから取得可能な情報のみを用いても検知できるような特殊なデータ変換をおこなう。また、仮想化ソフトウェアで前処理を行ってからクラウドで後処理をおこなうことでデータ転送量を削減し、速やかなマルウェア検出を可能にする。これにより、高効率・高検出率で回避不可能な強力なバンキングマルウェア対策を可能にする。また将来的な深層学習と仮想化技術を融合したマルウェア高精度検出という新しい体系の一般化に向けた道筋をつける。

Outline of Final Research Achievements

In this study, we conducted research on countermeasures against banking malware by integrating deep learning and virtualization technologies. We studied a method that can classify variants with high accuracy by deep learning, taking advantage of the fact that banking malware has many variants. We also studied a technique that can generate and detect malware images from binary-level data that can be obtained by virtualization technology, with the aim of integrating deep learning technique with virtualization technology. We explored various models and parameter combinations for deep learning, including the use of labels for confidentiality measures, and found that the latest models provide high classification accuracy with a low degree of transition learning.

Academic Significance and Societal Importance of the Research Achievements

近年は非常に多数のマルウェアが登場しており、実際に様々なセキュリティ上の被害が継続的に発生し続けているのが現状である。本研究では、最新の深層学習技術と仮想化技術を融合することで、高い精度でマルウェアを検知することができて、かつマルウェアが回避できないシステムの構築に向けた基礎研究を実施した。この技術を発展させることにより、将来的にマルウェアによる被害を大幅に低減できるシステムが実用化されることが見込まれる。

Report

(3 results)
  • 2021 Annual Research Report   Final Research Report ( PDF )
  • 2020 Research-status Report
  • Research Products

    (2 results)

All 2022

All Journal Article (1 results) (of which Peer Reviewed: 1 results) Presentation (1 results) (of which Int'l Joint Research: 1 results)

  • [Journal Article] Exploring Optimal Deep Learning Models for Image-based Malware Variant Classification2022

    • Author(s)
      Rikima Mitsuhashi and Takahiro Shinagawa
    • Journal Title

      In Proceedings of the IEEE 46th Annual Computers, Software, and Applications Conference (COMPSAC 2022)

      Volume: -

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed
  • [Presentation] Deriving Optimal Deep Learning Models for Image-based Malware Classification2022

    • Author(s)
      Rikima Mitsuhashi and Takahiro Shinagawa.
    • Organizer
      37th ACM/SIGAPP Symposium On Applied Computing
    • Related Report
      2021 Annual Research Report
    • Int'l Joint Research

URL: 

Published: 2020-08-03   Modified: 2023-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi