It Seems Twitter’s Troll Filter Works Better Than Removing Abusive Posts

Twitter has been trying to resolve its troll issue since 2006 and it has resorted to all kinds of things, from deleting accounts that are exhibiting offensive behavior to providing account holders with tools to mute other users. In a recent statement, however, Twitter revealed that it found a better way to handle trolls and it’s not be trying to get rid of them completely.

In a recent blog post, Twitter provided the initial results of a new filtering system that it’s testing out to scan for bad behaviors among users and then make them less visible. This is similar to how Google ranks websites based on trustworthiness, with the least trustworthy being placed way past the first page of its search results.

“In our early testing in markets around the world, we’ve already seen this new approach have a positive impact, resulting in a 4% drop in abuse reports from search and 8% fewer abuse reports from conversations. That means fewer people are seeing Tweets that disrupt their experience on Twitter,” the post reads.

These results come after Twitter CEO Jack Dorsey promised that the conversational health would improve on Twitter a few months back, Mashable reports. It would seem that the company has been hard at work in actually making this happen.

By basing its filter on behavioral patterns instead of focusing on language, Twitter is better able to distinguish the real trolls from those who are just exchanging friendly banter. The platform is basically interested in disruptive trolls who only engage with others with the expressed purpose of sowing chaos.

Once these trolls have been identified, they’ll become less visible and their offensive Tweets will have less impact. Not surprisingly, the trolls have come out to express their outrage at being branded as such and for Twitter’s supposed censorship rampage.