The Dark Web is the unparalleled go-to alternative for entities seeking enhanced anonymity online. While serving as a safe space for targeted individuals, it simultaneously attracts actors with illicit intentions. To further mask the activity of criminal businesses on the Dark Web, the usage of cryptocurrencies, especially Bitcoin, has become all the more prevalent thanks to their decentralized pseudonymous design. A substantial amount of research has been performed on the abuse of Bitcoin and the Dark Web, respectively – though methods to counteract their synergy remain limited.

In this thesis, we construct and apply a methodology for collecting Bitcoin address data, in connection to the website on which they are present, from the Dark Web. By extracting and traversing a total of 196,591 onion domains, we get 42,696 matches for Bitcoin addresses. Then, we retrieve transaction data for every valid Bitcoin address and consolidate the acquired information into a complete dataset. With this, we also compile profiles for select Dark Web marketplaces to get an overview of how these services are operated. 

Based on our findings, we conclude that the utilization of a web crawler, together with the Ahmia search engine for seed addresses, is effective for collecting Bitcoin-related data from the Dark Web; although, its efficiency is significantly enhanced by the introduction of multithreading. Furthermore, we infer that most Dark Web addresses do not contain Bitcoin addresses, though ones that do, generally have a single associated address, typically as their service's sole payment option. Lastly, we deduce that the most widespread Dark Web marketplace topics, according to our data, are stolen funds (such as credit cards or PayPal accounts) and abusive content (including rape and child pornography).