Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

You'd think they have the world's biggest dataset of scam posts too, with people constantly labeling posts for them (including law enforcement).


The problem is that scam posts are frequently just copied legitimate posts. The only thing we found that worked was to force communication to go through a messaging proxy to run classifiers on those messages.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: