Online Program Home
My Program

Abstract Details

Activity Number: 291 - Statistical Applications in Forensic Evidence
Type: Topic Contributed
Date/Time: Tuesday, July 31, 2018 : 8:30 AM to 10:20 AM
Sponsor: Advisory Committee on Forensic Science
Abstract #328662 Presentation
Title: Approaches to Matching Darknet Market Seller
Author(s): Xiao Hui Tai*
Companies: Carnegie Mellon University
Keywords: forensic matching; record linkage; classification

Online anonymous marketplaces or darknet marketplaces are commercial marketplaces that run on the dark web (parts of the web accessible only through anonymizing software such as Tor). Buyers and sellers preserve anonymity by using cryptocurrencies and PGP. As a result these marketplaces are used primarily for illegal activities such as sales of drugs. There are many such marketplaces, and sellers can operate accounts on different marketplaces, or multiple accounts on the same marketplace. We explore methods to match these accounts, to predict which of them belong to the same sellers. This is analogous to the forensic matching of DNA, cartridge cases, and so forth. In this talk we describe various approaches to matching, beginning with supervised methods, where labels are generated using PGP keys. Some of these labels might be erroneous, and we try to fix these, for example using an active learning approach. We also consider the possibility of using semi-supervised or unsupervised methods.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program