Twitter changes algorithm to reduce trolling

Twitter announced Tuesday significant changes to how it battles abusive users of its service.

The company has long been criticized for giving a platform to so-called “trolls”, users that willfully harass others. While trolling is often harmless, some trolls have targeted users with campaigns of death threats and other abusive activity.

The company said it has revised its algorithm to identify abusive behavior. Once a user is identified as a harassing troll, the user’s tweets would be pushed downward in Twitter responses and searches no matter the content of the tweet.

The concept behind the algorithm redesign is to steeply reduce the visibility of these trolls.

“We’re tackling issues of behaviors that distort and detract from the public conversation in those areas by integrating new behavioral signals into how tweets are presented,” Twitter executives Del Harvey and David Gasca said in an announcement.

“By using new tools to address this conduct from a behavioral perspective, we’re able to improve the health of the conversation, and everyone’s experience on Twitter, without waiting for people who use Twitter to report potential issues to us.”

Since it was founded in 2006, Twitter has been known for giving a platform to nearly unrestricted public discourse. While the lack of censorship has been a key pillar for the service’s popularity, there have been issues with hate groups and violent extremists using the service to spread their message.

Twitter CEO Jack Dorsey said the new algorithm is the latest step toward reducing the reach of problematic accounts.

“Healthier conversations is our goal, and how we will measure our progress,” Dorsey tweeted Tuesday.

The company said that in early testing around the world, the new algorithm has resulted in an 8 percent decrease in the number of abuse reports stemming from user interactions on the platform.