FWIW we ended up tracking TOR exit nodes on Delicious, and then we used "account created on tor" as a flag to the bayesian.
Were you able to get a bayesian algorithm to learn whether specific users were likely to be spammers or trolls? (As opposed to specific posts.) I'm guessing this isn't easy to do, since I get spammed on other services on a regular basis (e.g. Twitter spammers will setup new accounts and spam on a regular basis).
i believe it's possible to run a node and just query it for the exit nodes, but I'm not sure.