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

Why are you assuming there's anyone who could help? Machine learning is used heavily in anti-fraud systems. Machine learning is also relatively opaque—a spam filter can't be prodded to "explain its thinking." I suspect there's frequently not a single human being in the company who could pull up the raw score vector fed to their anti-fraud model and use it to tell you what exactly made you look bad.


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

Search: