SYSML: StYlometry with Structure and Multitask Learning: Implications for Darknet Forum Migrant Analysis

• DOI: 10.18653/v1/2021.emnlp-main.548
• Publication Date: 2021-12-16T22:56:42Z
• Authors: Pranav Maneriker; Yuntian He; Srinivasan Partha...
• Keywords: Stylometry; Lift (data mining); Rank (graph theory)

Darknet market forums are frequently used to exchange illegal goods and services between parties who use encryption conceal their identities. The Tor network is host these markets, which guarantees additional anonymization from IP location tracking, making it challenging link across malicious users using multiple accounts (sybils). Additionally, migrate new when one closed further increasing the difficulty of linking forums. We develop a novel stylometry-based multitask learning approach for natural language model interactions graph embeddings construct low-dimensional representations short episodes user activity authorship attribution. provide comprehensive evaluation our methods four different darknet demonstrating its efficacy over state-of-the-art, with lift up 2.5X on Mean Retrieval Rank 2X Recall@10.