Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Cal researchers build AI that can flag Twitter Propaganda (wired.com)
10 points by sahaskatta on Oct 31, 2017 | hide | past | favorite | 1 comment


"As fall semester kicked up, RoBhat hand-picked 100 Twitter accounts with automated behavior to serve as “ground truth” data to train their model. They picked accounts with several red flags: ones that joined the site, say, a month prior but had tweeted 10,000 times, or ones that were followed by thousands of other suspected bots. (“I no longer feel bad about how few Twitter followers I have,” quips Bhat, follower count: 1,250.) They then added those accounts’ followers into the “ground truth” set as well. They needed a large number for their machine to analyze—6,000 in all."

Seems a little bit dubious to label followers as ground truth data.




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

Search: