A Topic-Based Unsupervised Learning Approach for Online Underground Market Exploration

Cyber fraud has become a lucrative form of illicit business by leveraging the Internet as a communication channel and as a result, causes significant losses to the economy. Criminals in the cyber fraud underground economy use online underground markets and other forms of social media to exchange and...

Full description

Saved in:
Bibliographic Details
Published in:IEEE ... International Conference on Trust, Security and Privacy in Computing and Communications (Online) pp. 208 - 215
Main Authors: Huang, Shin-Ying, Ban, Tao
Format: Conference Proceeding
Language:English
Japanese
Published: IEEE 01.08.2019
Subjects:
black market
clustering
Computer crime
Credit cards
cybersecurity
Ecosystems
Feature extraction
growing hierarchical self-organizing map
Self-organizing feature maps
social network analysis
Social networking (online)
Text mining
threat intelligence
ISSN:2324-9013
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Cyber fraud has become a lucrative form of illicit business by leveraging the Internet as a communication channel and as a result, causes significant losses to the economy. Criminals in the cyber fraud underground economy use online underground markets and other forms of social media to exchange and trade illegitimate information. Due to the high variability in the marketplaces and actors therein, analyzing these underground markets is challenging. To understand more about the underground economy of cyber fraud and its actors, we propose a topic-based hierarchical self-organizing map, which can well represent and visualize actors' similarity and thus, uncover their roles in the underground markets. We compare the proposed method with a topic-based social network analysis method for identifying the key users and their roles in the cyber fraud value chain. Experiments conducted on data from several online underground markets suggest that the proposed method can aid in identifying key actors in terms of roles, influence levels, and their social relationships.
ISSN:2324-9013
DOI:10.1109/TrustCom/BigDataSE.2019.00036