SEC and FINRA invest a lot in Data Analytics. Go to https://technology.finra.org/tech.html and you'll see that they use the same tools as the internet industry.

I'm not even sure you need sophisticated data analysis.

The two people who did the biggest trades:

- weren't involved in the finance industry (a school teacher and an equipment leasing business)

- had never traded options before

- did the brokerage know-your-client process to get approved for options 1-2 weeks before quarterly earnings came out

- profited $884,000 and $175,000

A SQL query looking for people who made > 100K within 2 weeks of opening their account would catch these people, and probably have relatively few false positives.