I signed up with a vpn and got banned for life after a single nonsense tweet about not liking the feed and following 4-5 famous people. I don’t think bot detection techniques are very robust

> This methodology likely undercounts spam and fake accounts, but almost never includes false positives (i.e. claiming an account is fake when it isn’t).

In other words, their model performs well on their training set, and they don’t acknowledge that it may be over fitted or mislabeled, and they hand wave mistakes