Description
In the first domain, we develop tools and techniques that can be used separately and in conjunction to group Backpage sex ads by their true author (and not the claimed author in the ad). Sites for online classified ads selling sex are widely used by human traffickers to support their pernicious business. The sheer quantity of ads makes manual exploration and analysis unscalable. In addition, discerning whether an ad is advertising a trafficked victim or an independent sex worker is a very difficult task. Very little concrete ground truth (i.e., ads definitively known to be posted by a trafficker) exists in this space. In the first chapter of this dissertation, we develop a machine learning classifier that uses stylometry to distinguish between ads posted by the same vs. different authors with 90% TPR and 1% FPR. We also design a linking technique that takes advantage of leakages from the Bitcoin mempool, blockchain and sex ad site, to link a subset of sex ads to Bitcoin public wallets and transactions. Finally, we demonstrate via a 4-week proof of concept using Backpage as the sex ad site, how an analyst can use these automated approaches to potentially find human traffickers.
In the second domain, we develop machine learning tools to classify and extract information from cyber black-market forums. Underground forums are widely used by criminals to buy and sell a host of stolen items, datasets, resources, and criminal services. These forums contain important resources for understanding cybercrime. However, the number of forums, their size, and the domain expertise required to understand the markets makes manual exploration of these forums unscalable. In the second chapter of this dissertation, we propose an automated, top-down approach for analyzing underground forums. Our approach uses natural language processing and machine learning to automatically generate high-level information about underground forums, first identifying posts related to transactions, and then extracting products and prices. We also demonstrate, via a pair of case studies, how an analyst can use these automated approaches to investigate other categories of products and transactions. We use eight distinct forums to assess our tools: Antichat, Blackhat World, Carders, Darkode, Hack Forums, Hell, L33tCrew and Nulled. Our automated approach is fast and accurate, achieving over 80% accuracy in detecting post category, product, and prices.
In the third domain, we develop a set of features for a principal component analysis (PCA) based anomaly detection system to extract producers (those actively abusing children) from the full set of users on Tor CSAM forums. These forums are visited by tens of thousands of pedophiles daily. The sheer quantity of users and posts make manual exploration and analysis unscalable. In the final chapter of this dissertation, we demonstrate how to extract producers from unlabeled, public forum data. We use four distinct forums to assess our tools; these forums remain unnamed to protect law enforcement investigative efforts.
We have released our code written for the first two domains, as well as the proof of concept data from the first domain, and a sub-set of the labeled data from the second domain, allowing replication of our results.