Gibran Gomez

Multiple works have leveraged the public Bitcoin ledger to estimate the revenue cybercriminals obtain from their victims. Estimations focusing on the same target often do not agree, due to the use of different methodologies, seed addresses, and time periods. These factors make it challenging to understand the impact of their methodological differences. Furthermore, they underestimate the revenue due to the (lack of) coverage on the target's payment addresses, but how large this impact remains unknown. In this work, we perform the first systematic analysis on the estimation of cybercrime bitcoin revenue. We implement a tool that can replicate the different estimation methodologies. Using our tool we can quantify, in a controlled setting, the impact of the different methodology steps. In contrast to what is widely believed, we show that the revenue is not always underestimated. There exist methodologies that can introduce huge overestimation. We collect 30,424 payment addresses and use them to compare the financial impact of 6 cybercrimes (ransomware, clippers, sextortion, Ponzi schemes, giveaway scams, exchange scams) and of 141 cybercriminal groups. We observe that the popular multi-input clustering fails to discover addresses for 40% of groups. We quantify, for the first time, the impact of the (lack of) coverage on the estimation. For this, we propose two techniques to achieve high coverage, possibly nearly complete, on the DeadBolt server ransomware. Our expanded coverage enables estimating DeadBolt's revenue at $2.47M, 39 times higher than the estimation using two popular Internet scan engines.

Malware abuses TLS to encrypt its malicious traffic, preventing examination by content signatures and deep packet inspection. Network detection of malicious TLS flows is important, but it is a challenging problem. Prior works have proposed supervised machine learning detectors using TLS features. However, by trying to represent all malicious traffic, supervised binary detectors produce models that are too loose, thus introducing errors. Furthermore, they do not distinguish flows generated by different malware. On the other hand, supervised multiclass detectors produce tighter models and can classify flows by the malware family but require family labels, which are not available for many samples. To address these limitations, this work proposes a novel unsupervised approach to detect and cluster malicious TLS flows. Our approach takes input network traces from sandboxes. It clusters similar TLS flows using 90 features that capture properties of the TLS client, TLS server, certificate, and encrypted payload and uses the clusters to build an unsupervised detector that can assign a malicious flow to the cluster it belongs to, or determine if it is benign. We evaluate our approach using 972K traces from a commercial sandbox and 35M TLS flows from a research network. Our clustering shows very high precision and recall with an F1 score of 0.993. We compare our unsupervised detector with two state-of-the-art approaches, showing that it outperforms both. The false detection rate of our detector is 0.032 measured over four months of traffic.

Malware abuses TLS to encrypt its malicious traffic, preventing examination by content signatures and deep packet inspection. Network detection of malicious TLS flows is an important, but challenging, problem. Prior works have proposed supervised machine learning detectors using TLS features. However, by trying to represent all malicious traffic, supervised binary detectors produce models that are too loose, thus introducing errors. Furthermore, they do not distinguish flows generated by different malware. On the other hand, supervised multi-class detectors produce tighter models and can classify flows by malware family, but require family labels, which are not available for many samples. To address these limitations, this work proposes a novel unsupervised approach to detect and cluster malicious TLS flows. Our approach takes as input network traces from sandboxes. It clusters similar TLS flows using 90 features that capture properties of the TLS client, TLS server, certificate, and encrypted payload; and uses the clusters to build an unsupervised detector that can assign a malicious flow to the cluster it belongs to, or determine it is benign. We evaluate our approach using 972K traces from a commercial sandbox and 35M TLS flows from a research network. Our clustering shows very high precision and recall with an F1 score of 0.993. We compare our unsupervised detector with two state-of-the-art approaches, showing that it outperforms both. The false detection rate of our detector is 0.032% measured over four months of traffic.

Cybercriminals often leverage Bitcoin for their illicit activities. In this work, we propose back-and-forth exploration, a novel automated Bitcoin transaction tracing technique to identify cybercrime financial relationships. Given seed addresses belonging to a cybercrime campaign, it outputs a transaction graph, and identifies paths corresponding to relationships between the campaign under study and external services and other cybercrime campaigns. Back-and-forth exploration provides two key contributions. First, it explores both forward and backwards, instead of only forward as done by prior work, enabling the discovery of relationships that cannot be found by only exploring forward (e.g., deposits from clients of a mixer). Second, it prevents graph explosion by combining a tagging database with a machine learning classifier for identifying addresses belonging to exchanges. We evaluate back-and-forth exploration on 30 malware families. We build oracles for 4 families using Bitcoin for C&C and use them to demonstrate that back-and-forth exploration identifies 13 C&C signaling addresses missed by prior work, 8 of which are fundamentally missed by forward-only explorations. Our approach uncovers a wealth of services used by the malware including 44 exchanges, 11 gambling sites, 5 payment service providers, 4 underground markets, 4 mining pools, and 2 mixers. In 4 families, the relations include new attribution points missed by forward-only explorations. It also identifies relationships between the malware families and other cybercrime campaigns, highlighting how some malware operators participate in a variety of cybercriminal activities.