계층적인 잠재 표현 기반의 사이버 범죄 신조어 자동 탐지 프레임워크 = Hierarchical Latent Representation-based Framework for Automatic Detection of Cybercrime Slang

다국어 초록 (Multilingual Abstract)

Cybercriminals constantly produce and use slang by adding criminal meanings to existing words or replacing them with similar words for communication. Continuous monitoring and manual work are required to respond to this, and a large amount of labeled training data is required when using deep learning. However, the ability to collect a large amount of training data is limited because direct labeling by a person requires a lot of time and money and proceeds secretly due to the nature of cybercrime. Thus, we develop a framework based on an autoencoder and propose a method to effectively detect contextual cybercrime slang and neologisms through hierarchical latent vector similarity comparisons to address these limitations. Experiments using a cybercrime post dataset showed that the framework had an accuracy of up to 99.1% at a similarity threshold of 0.5.

참고문헌 (Reference)

1 박준영 ; 채명수 ; 정성관, "실시간 범죄 예측을 위한 랜덤포레스트 알고리즘 기반의 범죄 유형 분류모델 및 모니터링 인터페이스 디자인 요소 제안" 한국정보과학회 22 (22): 455-460, 2016

2 문현지 ; 박현민 ; 김기범, "보이스피싱 범죄조직의 은어 사용과 AI 기술을 적용한 수사상 활용방안 연구" 경찰대학 21 (21): 137-160, 2021

3 L. Cheng, "Unsupervised Cyberbullying Detection via Time-Informed Gaussian Mixture Model"

4 L. Ranaldi, "THE DARK SIDE OF THE LANGUAGE : PRE-TRAINED TRANSFORMERS IN THE DARKNET"

5 P. Vincent, "Stacked Denoising Autoencoders : Learning Useful Representations in a Deep Network with a Local Denoising Criterion" 11 (11): 3371-3408, 2010

6 "Smart Policing Big Data Platform"

7 "Smart Policing Big Data Platform"

8 "Smart Policing Big Data Platform"

9 "Smart Policing Big Data Platform"

10 최은정 ; 이수련 ; 권혜민 ; 김명주 ; 이인수 ; 이승훈, "SNS 빅데이터 및 검색포털 트렌드와 마약류 사건 통계간의 비교 및 의미분석 연구" 한국디지털정책학회 19 (19): 231-238, 2021

11 X. Xiaofen, "SAE+ LSTM: A New framework for emotion recognition from multi channel EEG" 13 (13): 2019

12 X. Shiqiang, "Re search on Argument Text Clustering Method Based on Autoencoder" 21 (21): 181-192, 2021

13 Z. Zhao, "LSHWE : improving similarity-based word embedding with locality sensitive hashing for cyberbullying detection" 8 (8): 328-339, 2016

14 "KOREAN NATIONAL POLICE AGENCY"

15 B. Yoshuna, "Generalized Denoising Auto-Encoders as Generative Models" 26 : 2013

16 G. Pengfei, "Dual Adversarial Autoencoders for Clustering" 21 (21): 1417-1424, 2020

17 H. AL-Saif, "Detecting and Classifying Crimes from Arabic Twitter Posts using Text Mining Techniques" 9 (9): 377-387, 2018

18 A. Muneer, "Cyberbullying Detection on Social Media Using Stacking Ensemble Learning and Enhanced BERT" 14 (14): 467-, 2023

19 R. Zhao, "Cyberbullying Detection Based on Semantic-Enhanced Marginalized Denoising Auto-Encoder" 2020

20 H. Zhou, "Cross-Lingual Sentiment Classification Based on Denoising Autoencoder" 496 : 181-192, 2014

21 S. Sathyadevan, "Crime analysis and prediction using data mining" 9 (9): 406-412, 2014

22 Z. Beiji, "Crime Hotspot Detection and Monitoring Using Video Based Event Modeling and Mapping Techniques" 10 (10): 962-969, 2017

23 I. Jayaweera, "Crime Analytics : Analysis of Crimes Through Newspaper Articles" 277-282, 2015

24 W. Yasi, "Auto-encoder based dimensionality reduction" 19 : 232-242, 2016

25 D. P. Kingma, "Auto-Encoding Variational Bayes"

26 S. Lal, "Analysis and Classification of Crime Tweets" 167 : 1911-1919, 2022

27 K. Zahra, "Analysis and Classification of Crime Tweets" 11-16, 2018

28 A. Makhzani, "Adversarial Autoencoders"

29 "AIHub"

30 L. Ruizhe, "A stable variational autoencoder for text modelling" 13 (13): 594-599, 2019

31 G. Deepak, "A knowledge centric hybridized approach for crime classification incorporating deep bi-LSTM neural network" 80 (80): 28061-28085, 2021

32 K. B. Sundhara Kumar, "A Novel Hybrid RNN-ELM Architecture for Crime Classification" 876-882, 2020

33 E. A. Emon, "A Deep Learning Approach to Detect Abusive Bengali Text" 1-5, 2019

34 B. Kim, "2022 Drug Crime White Paper" Supreme (Public) Prosecutors' Office 4-, 2023