Bitcoin offers many new opportunities and challenges with its pseudonymity and open source nature. One of the challenges is performing taint analysis in order to follow coins that originated from criminal activities. Due to a large number of nodes and the complexity of the Bitcoin transaction graph, methods for the performance of taint analysis have been developed. In this master’s thesis, existing methods were implemented and furthermore a new method called COMB was proposed. A database that supports running these methods was put together. For the testing purpose, two data sets of starting transaction outputs were prepared. After executing all methods on the data sets and analysis of the results, it was concluded that all methods have pros and cons. The intersections of graphs produced by different algorithms from the same starting inputs were analyzed, because they contain transactions with a higher probability of being connected to the starting transaction out- put. Another database with off-chain data that can be used in implemented methods was developed. Even with a relatively small database, we were able to reach some known transactions with implemented methods, showing the big potential of this technique.
bitcoin_blk_to_rocksdb- Source code (C++) for transforming bitcoin.blkfiles toRocksDBdatabasealgorithms- Source code (C++) of algorithms for different tainting methodsMasters_thesis.pdf- Master's thesis (mostly in Slovene language, abstract is available in English)
I analysed data with Python programming language with following libraries:
networkxstatisticscommunitypowerlawmatplotlibscipy