We investigate to which degree one could trace Bitcoin transactions and characterize purchasing behavior of online anonymous marketplaces by exploiting side channels. Using a list of addresses found by the FBI on Silk Road servers, and information on the marketplace's official guides, we infer the role played by each address in the list and classify them based on heuristics. We then attempt to trace Bitcoin transactions and show that the anonymity set size is greatly reduced using product review data and the address classification performed on the previous step. Finally, using clustering techniques based on transaction graph analysis, we assign addresses into user wallets, then group these wallets together based on spending patterns, to be able to characterize purchasing behavior.
Identifer | oai:union.ndltd.org:cmu.edu/oai:repository.cmu.edu:theses-1134 |
Date | 01 December 2017 |
Creators | Garcia, Eugene Lemuel R. |
Publisher | Research Showcase @ CMU |
Source Sets | Carnegie Mellon University |
Detected Language | English |
Type | text |
Format | application/pdf |
Source | Theses |
Rights | http://creativecommons.org/licenses/by-nc/4.0/ |
Page generated in 0.0143 seconds