Researchers Have Figured Out Troll Behavior and Are Trying to Fix It

Now, if you'll excuse me, I have to anger the internet.
This article is over 9 years old and may contain outdated information

Recommended Videos

Some researchers at Cornell University figured out how to identify internet trolls and are working on a way to deal with them. Loki is amused.

Justin Cheng, Cristian Danescu-Niculescu-Mizil, and Jure Leskovec submitted a paper earlier this month, explaining their findings after an 18-month long study. In cooperation with Disqus and partially funded by Google—two parties with a sizable troll problem—the study looked at the habits of banned commenters on three websites: cnn.com (news), breitbart.com (politics), and ign.com (gaming).

The researchers were looking for ways to anticipate trolls, who were referred to as FBU (future banned users), by studying their habits. The study examined over 10,000 FBUs and found that their commenting life began with a lower level of literacy or clarity in comparison to the rest of the group that they were conversing with. They were generally semi-literate, provocative, and fairly persistent before they eventually got banned.

The study noticed that on CNN, the trolls often initiated new posts or sub-threads. On Breitbart and IGN, they usually commented on existing threads.

The study also found that communities less tolerant of trolling were actually more likely to foster trolls. At the same time, patient communities would become less tolerant over time the longer FBUs remained in the community and eventually ban them as well. The more troll-intolerant the community, the more likely there will be a troll. The more likely one is a troll, the more likely one is banned. The more banning there is, the more intolerant a community looks.

And thus, the trollouroboros is born, and we are all doomed. They tested out an automated analysis system to identify BFUs early and deal with the vicious cycle, but 1 in 5 users were misclassified, so that still needs some work.

Can an automatic system really one day ban trolls? Would you trust it? Would it be worthy enough to even lift the mighty ban-hammer? What do you think?

(via The Stack)

—Please make note of The Mary Sue’s general comment policy.—

Do you follow The Mary Sue on Twitter, Facebook, Tumblr, Pinterest, & Google +?


The Mary Sue is supported by our audience. When you purchase through links on our site, we may earn a small affiliate commission. Learn more about our Affiliate Policy