GRU 기반 단축 URL 판별 기법을 적용한하이브리드 피싱 사이트 탐지 시스템 = Hybrid phishing site detection system with GRU-based shortened URL determination technique

국문 초록 (Abstract)

경찰청 통계자료에 따르면 코로나19 이후 문자 또는 메신저를 이용한 스미싱(Smishing) 범죄가 급증하였다. 또한 정부 기관에접수된 공공기관 사칭 건수의 대부분이 백신접종 및 보상 관련하여 가짜 URL(Uniform Resource Locator)을 클릭하도록 유도하는 수법이 다수 사용되었다. 주로 URL의 정보를 숨긴 단축 URL을 사용하며 탐지할 때 URL 기반 탐지방법은 URL의 정보를 숨기면 제대로 탐지할 수 없고, 콘텐츠 기반 탐지 방법은 탐지 속도가 느리고 많은 자원을 사용한다. 이에 본 논문에서는 GRU(GatedRecurrent Units)를 이용한 단축 URL을 판별하는 과정을 통해 일반 URL일 때 transformer를 통한 URL 기반 탐지, 단축 URL일때 XGBoost를 이용한 콘텐츠 기반 탐지하는 시스템을 제안한다. 제안한 탐지 시스템의 F1-Score는 94.86이었고, 처리시간은 평균 5.4초가 소요되었다.

다국어 초록 (Multilingual Abstract)

According to statistics from the National Police Agency, smishing crimes using texts or messengers haveincreased dramatically since COVID-19. In addition, most of the cases of impersonation of public institutionsreported to agency were related to vaccination and reward, and many methods were used to trick people intoclicking on fake URLs (Uniform Resource Locators). When detecting them, URL-based detection methods cannotdetect them properly if the information of the URL is hidden, and content-based detection methods are slow anduse a lot of resources. In this paper, we propose a system for URL-based detection using transformer for regularURLs and content-based detection using XGBoost for shortened URLs through the process of determiningshortened URLs using GRU(Gated Recurrent Units). The F1-Score of the proposed detection system was 94.86, andits average processing time was 5.4 seconds.

참고문헌 (Reference)

• 1 dmlc XGBoost, "XGBoost Documentation"
• 2 A. Hannousse, "Web page phishing detection"
• 3 Julio Lamas Piñeiro, "Web architecture for URL-based phishing detection based on Random Forest, Classification Trees, and Support Vector Machine" IBERAMIA: Sociedad Iberoamericana de Inteligencia Artificial 25 (25): 107-121, 2022
• 4 Y. Kim, "Short URLs Verification Approach for Phishing Site Detection Improvement" 80-81, 2022
• 5 Jordan. M. I, "Serial order: a parallel distributed processing approach" 1986
• 6 Sumitra Das Guptta, "Modeling Hybrid Feature-Based Phishing Websites Detection Using Machine Learning Techniques" Springer Science and Business Media LLC 2022
• 7 S. Hochreiter, "LONG SHORT-TERM MEMORY" 9 (9): 1735-1780, 1997
• 8 Jerome H. Friedman, "Greedy function approximation: A gradient boosting machine." Institute of Mathematical Statistics 29 (29): 1189-1232, 2001
• 9 J. Chung, "Empirical evaluation of gated recurrent neural networks on sequence modeling"
• 10 Korean National Police Agency, "Cyber Criminal Trend for 2022"
• 11 U. Ozker, "Content Based Phishing Detection with Machine Learning" 1-6, 2020
• 12 A. Vaswani, "Attention is all you need" 30 : 2017
• 13 Mehmet Korkmaz, "A Hybrid Phishing Detection System Using Deep Learning-based URL and Content Analysis" Kaunas University of Technology (KTU) 28 (28): 80-89, 2022