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

Research for Problems in Information Security Caused by Application of Machine Learning

Research Project

Project/Area Number 18K11248
Research Category

Grant-in-Aid for Scientific Research (C)

Allocation TypeMulti-year Fund
Section一般
Review Section Basic Section 60050:Software-related
Research InstitutionTokyo University of Technology

Principal Investigator

Uda Ryuya  東京工科大学, コンピュータサイエンス学部, 准教授 (50350509)

Co-Investigator(Kenkyū-buntansha) 柴田 千尋  法政大学, 理工学部, 准教授 (00633299)
Project Period (FY) 2018-04-01 – 2023-03-31
Project Status Completed (Fiscal Year 2022)
Budget Amount *help
¥4,420,000 (Direct Cost: ¥3,400,000、Indirect Cost: ¥1,020,000)
Fiscal Year 2020: ¥260,000 (Direct Cost: ¥200,000、Indirect Cost: ¥60,000)
Fiscal Year 2019: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000)
Fiscal Year 2018: ¥3,250,000 (Direct Cost: ¥2,500,000、Indirect Cost: ¥750,000)
Keywords情報セキュリティ / 人工知能 / 機械学習 / 深層学習 / 標的型マルウェア / Adversarial Examples
Outline of Final Research Achievements

The greatest effort is Adversarial CAPTCHA which has strong removal resistance while keeping its visibility. Other researches are pointing out a problem in researches for detecting Cross Site Scripting with machine learning, recognizing unreadable numbers on license plates by machine learning, personal verification with any hand-written scripts by disassembling writing features and high accuracy malware detection by size compression with malicious features while reducing time for machine learning.

Academic Significance and Societal Importance of the Research Achievements

機械学習があらゆるものに利用されるようになり、情報セキュリティ分野のサービスやシステムにも利用されるようになってきた。一方、機械学習に詳しい者が情報セキュリティに詳しいとは限らず、また逆もしかりであるため、開発された技術に問題がある場合や、開発自体を断念してしまうこともあり得る。研究成果のAdversarial CAPTCHAは、万能と思われていた人工知能技術に一石を投じるものであったと言える。XSS検出技術における問題点の指摘や、大量の写真を使わずにナンバープレートの数字を読む技術は、通常とは異なる視点からの解を社会に与えられたと考えている。

Report

(6 results)
  • 2022 Annual Research Report   Final Research Report ( PDF )
  • 2021 Research-status Report
  • 2020 Research-status Report
  • 2019 Research-status Report
  • 2018 Research-status Report
  • Research Products

    (23 results)

All 2022 2021 2020 2019

All Journal Article (13 results) (of which Peer Reviewed: 8 results) Presentation (10 results) (of which Int'l Joint Research: 5 results)

  • [Journal Article] n-gram抽出と機械学習を用いた亜種マルウェア分類手法の提案と評価2022

    • Author(s)
      瀧口翔貴, 宇田隆哉
    • Journal Title

      情報処理学会論文誌

      Volume: 63 Pages: 1052-1071

    • Related Report
      2021 Research-status Report
    • Peer Reviewed
  • [Journal Article] 特徴の再訓練を必要としない変更可能な筆記2022

    • Author(s)
      釜石智史, 宇田隆哉
    • Journal Title

      情報処理学会論文誌

      Volume: 63 Pages: 1094-1114

    • Related Report
      2021 Research-status Report
    • Peer Reviewed
  • [Journal Article] n-gramによるマルウェア検出における機械学習を騙す良性ソフトウェア汚染2021

    • Author(s)
      宇田隆哉
    • Journal Title

      コンピュータセキュリティシンポジウム2021論文集

      Volume: 1 Pages: 623-630

    • NAID

      170000186032

    • Related Report
      2021 Research-status Report
  • [Journal Article] 不適切なデータセットや処理方法を用いた機械学習によるXSS攻撃検出研究の解説と精度の比較2021

    • Author(s)
      飯野和真, 宇田隆哉
    • Journal Title

      情報処理学会研究報告

      Volume: 2021-CSEC-92 Pages: 1-8

    • Related Report
      2020 Research-status Report
  • [Journal Article] Malware Subspecies Detection Method by Suffix Arrays and Machine Learning2021

    • Author(s)
      Kouhei Kita and Ryuya Uda
    • Journal Title

      Proceeding of the 55th Annual Conference on Information Sciences and Systems

      Volume: なし

    • Related Report
      2020 Research-status Report
    • Peer Reviewed
  • [Journal Article] Recognition of Digits on License Plate by RAISR with Changing Contrast Ratio2021

    • Author(s)
      Tomoya Suzuki and Ryuya Uda
    • Journal Title

      Proceeding of the 55th Annual Conference on Information Sciences and Systems

      Volume: なし

    • Related Report
      2020 Research-status Report
    • Peer Reviewed
  • [Journal Article] 圧縮サイズと比較コストを考慮したマルチN-gramによる亜種マルウェアの検出2020

    • Author(s)
      宇田隆哉
    • Journal Title

      情報処理学会コンピュータセキュリティシンポジウム2020論文集

      Volume: なし

    • NAID

      170000184009

    • Related Report
      2020 Research-status Report
  • [Journal Article] Comparison of Algorithms and Action Coordinates Sets in Detection of Slight Differences in Motions like Lock-Picking2020

    • Author(s)
      Masaki Shiraishi and Ryuya Uda
    • Journal Title

      Proceeding of the 5th South-East Europe Design Automation, Computer Engineering, Computer Networks and Social Media Conference

      Volume: なし

    • Related Report
      2020 Research-status Report
    • Peer Reviewed
  • [Journal Article] N-gram抽出法による亜種マルウェアの検出と攻撃耐性の考察2019

    • Author(s)
      宇田隆哉
    • Journal Title

      情報処理学会コンピュータセキュリティシンポジウム2019論文集

      Volume: 1 Pages: 515-522

    • NAID

      170000181045

    • Related Report
      2019 Research-status Report
  • [Journal Article] 畳込みニューラルネットワークに耐性のあるCAPTCHAの提案と評価2019

    • Author(s)
      阿座上知香, 柴田千尋, 宇田隆哉
    • Journal Title

      情報処理学会論文誌

      Volume: 60 Pages: 680-695

    • Related Report
      2018 Research-status Report
    • Peer Reviewed
  • [Journal Article] N-gram圧縮と深層学習を用いたマルウェア分類手法の提案2019

    • Author(s)
      瀧口翔貴, 宇田隆哉
    • Journal Title

      電子情報通信学会技術研究報告

      Volume: 118 Pages: 111-116

    • Related Report
      2018 Research-status Report
  • [Journal Article] Classification of XSS Attacks by Machine Learning with Frequency of Appearance and Co-Occurrence2019

    • Author(s)
      Sota Akaishi and Ryuya Uda
    • Journal Title

      The 53rd Annual Conference on Information Sciences and Systems

      Volume: なし Pages: 1-6

    • DOI

      10.1109/ciss.2019.8693047

    • Related Report
      2018 Research-status Report
    • Peer Reviewed
  • [Journal Article] Creation of Adversarial Examples with Keeping High Visual Performance2019

    • Author(s)
      Tomoka Azakami, Chihiro Shibata, Ryuya Uda and Toshiyuki Kinoshita
    • Journal Title

      IEEE 2nd International Conference on Information and Computer Technologies

      Volume: なし Pages: 52-56

    • DOI

      10.1109/infoct.2019.8710918

    • Related Report
      2018 Research-status Report
    • Peer Reviewed
  • [Presentation] n-gramによるマルウェア検出における機械学習を騙す良性ソフトウェア汚染2021

    • Author(s)
      宇田隆哉
    • Organizer
      情報処理学会, コンピュータセキュリティシンポジウム2021
    • Related Report
      2021 Research-status Report
  • [Presentation] 不適切なデータセットや処理方法を用いた機械学習によるXSS攻撃検出研究の解説と精度の比較2021

    • Author(s)
      飯野和真
    • Organizer
      情報処理学会コンピュータセキュリティ研究会
    • Related Report
      2020 Research-status Report
  • [Presentation] Malware Subspecies Detection Method by Suffix Arrays and Machine Learning2021

    • Author(s)
      Kouhei Kita
    • Organizer
      55th Annual Conference on Information Sciences and Systems
    • Related Report
      2020 Research-status Report
    • Int'l Joint Research
  • [Presentation] Recognition of Digits on License Plate by RAISR with Changing Contrast Ratio2021

    • Author(s)
      Tomoya Suzuki
    • Organizer
      55th Annual Conference on Information Sciences and Systems
    • Related Report
      2020 Research-status Report
    • Int'l Joint Research
  • [Presentation] 圧縮サイズと比較コストを考慮したマルチN-gramによる亜種マルウェアの検出2020

    • Author(s)
      宇田隆哉
    • Organizer
      情報処理学会コンピュータセキュリティシンポジウム2020
    • Related Report
      2020 Research-status Report
  • [Presentation] Comparison of Algorithms and Action Coordinates Sets in Detection of Slight Differences in Motions like Lock-Picking2020

    • Author(s)
      Masaki Shiraishi
    • Organizer
      2nd International Workshop on Security and Reliability of IoT Systems
    • Related Report
      2020 Research-status Report
    • Int'l Joint Research
  • [Presentation] N-gram抽出法による亜種マルウェアの検出と攻撃耐性の考察2019

    • Author(s)
      宇田隆哉
    • Organizer
      情報処理学会コンピュータセキュリティシンポジウム2019
    • Related Report
      2019 Research-status Report
  • [Presentation] N-gram圧縮と深層学習を用いたマルウェア分類手法の提案2019

    • Author(s)
      瀧口翔貴,宇田隆哉
    • Organizer
      電子情報通信学会
    • Related Report
      2018 Research-status Report
  • [Presentation] Classification of XSS Attacks by Machine Learning with Frequency of Appearance and Co-Occurrence2019

    • Author(s)
      Sota Akaishi and Ryuya Uda
    • Organizer
      IEEE
    • Related Report
      2018 Research-status Report
    • Int'l Joint Research
  • [Presentation] Creation of Adversarial Examples with Keeping High Visual Performance2019

    • Author(s)
      Tomoka Azakami, Chihiro Shibata, Ryuya Uda and Toshiyuki Kinoshita
    • Organizer
      IEEE
    • Related Report
      2018 Research-status Report
    • Int'l Joint Research

URL: 

Published: 2018-04-23   Modified: 2024-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi