I have a few questions about Bitcoin fraud investigation.
What are the characteristics of fraudulent bitcoin transactions? For example, I've read some articles saying that a transaction that has only 1-2 inputs but comparatively too large number of output addresses (e.g. 60) can be considered suspicious because there's a chance that the transaction has been used for money laundering. I'd like to know which features I should look at in order to find fraudulent patterns in Bitcoin transaction data: transactions with the very small number of input counts but extremely large number of output counts, transactions with the extremely high number of input counts but small number of output counts, etc.
Which pattern would be more suspicious: a transaction that has 1 input count but 100 output counts OR a transaction that has 100 input counts but 1 output count?
Is there a way to extract geographic information of Bitcoin transactions? For example, there are some websites that show a world map that is color-scaled based on the number of transactions that have occurred in each country. I wonder how they get the geographic information of each transaction.