Twitter is currently experimenting with various and possible solutions to quash the toxicity so prevalent on its microblog network…
Social site Twitter continues to wrestle with effective methods for reducing abuse and bullying on its platform. And, it’s working on some possible solutions to that end.
Twitter Toxicity Reduction Solution Experiments Underway
Twitter has an ongoing problem with bully and toxic tweets, particularly when it comes to replies. (Especially so when sensitive topics are part of conversations.)
Kayvon Beykpour, the company’s head of product, told The Next Web at last week’s CES or Consumer Electronics Show one solution might be to downrank abusive replies with machine learning:
“I think increasingly, leveraging machine learning to try and model the behaviors that we think are most optimal for that area. So for example, we would like to show replies that are most likely to be replied to. That’s one attribute you might want to optimize for, not the only attribute by any means. You’ll want to deemphasize replies that are likely to be blocked or reported for abuse.”
Of course, the problem is that every solution breeds new problems. It’s very difficult to derank comments and replies that are on the borderline. For instance, a reply might be considered provocative but not necessarily abusive or bullying.