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

> Given Facebook's financial resources, it should be no problem to increase the size of the team working on the tool. Like any engineering problem, the problem can be broken down into smaller parts

And I worked for 4 years on one such small part (internally called UFAC), trying to help potential false positives of such a system.

As for classifier with a true positive rate of 99.99999%, I don't know much but I don't think it's possible. But if there's someone out there who might know, then they should say so.



No classification system can be perfect; we are in agreement there.

From the outside, it doesn't seem like a strictly data science problem. Instead, it seems like a design / product / brand expectation issue as well.




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

Search: