Finally, from a population of tree plots we can produce a generative network, in the form of a directed graph. Such a graph summarizes the sources and sinks in the risk profile. Such an approach can be used to map the user journey, too.
Other coins can be analyzed in a similar way. For instance, zCash (ZEC) figures for on-chain volume and transaction count reflect data for transparent transactions only. During May-June 2018, 9.1% of ZEC transactions were shielded. If the trend of increasing distrust in big business and government continues, the share of transactions of a private nature is likely to increase, making more important to understand where, when, and, how people spend their Zcash coins. In that scenario, we map the steps from different conversion points: what people do before using ZEC, and what they do after. We identify emerging merchants, and map their risk profiles. In this way we discover the unlikely places where people spend privacy-coin money.
Enter the Rabbit-hole: follow the money
Finding a payment system that both sides trust can be challenging. Inferential network analysis is also useful to find where are the trust asymmetries around the usage of privacy cryptocurrencies.
The likely places where people spend money
There are places where users with a high propensity of "infection" are attracted to by experienced criminals. Again, since surface web nodes are more resilient than hidden services, it is easier to find those places.
Fullzstores
Packages of data known as “fullz,” each of which typically includes a person’s name, date of birth, mother’s maiden name, Social Security number, and e-mail address and password are available in stores and forums online. We develop a better understanding of the new entrant persona by looking at their online behavior: what other online services they use when accessing those "dark" marketplaces from the regular web. We also apply semantics: Google queries such as "buy fullz with bitcoin" is a newbie search, experienced people use search engines such as Duckduckgo.