Innovation

Twitter hunts down trolls and cyberbullies with IBM Watson

The social media platform says it will use IBM Watson's cognitive computing powers to detect harassment on Twitter before it's been reported.

At the 2017 IBM InterConnect conference in Las Vegas, Twitter announced a partnership with IBM Watson to use the cognitive computing platform to better combat abuse on its site.

It's not too difficult to see how interactions on Twitter can lead to abuse. Just look at Microsoft's Tay, an AI-powered Twitter bot created in the character of a teenage girl, who, in less than 24 hours after conception, began uttering sexist and racist slurs—slurs that she had picked up based on human behavior on the platform.

These types of interactions have prompted the 11-year-old social media site to make strides to prevent harassment on the social media platform, with its vice president of data strategy, Chris Moody, calling it a "number one priority" for the company.

At IBM's InterConnect conference, Moody talked about a new strategy to weed out the bullies: Harness the power of IBM Watson.

Twitter wants to catch abusive behavior before it's been spotted and reported, said Moody in an earlier blog post. The platform would then restrict functionality for the perpetrator for a certain amount of time. If the problem persists, Twitter could block the user's account.

SEE: IBM Watson: The smart person's guide

"For example, this change could come into effect if an account is repeatedly Tweeting without solicitation at non-followers or engaging in patterns of abusive behavior that is in violation of the Twitter Rules," Moody wrote in the post.

With Watson, the subtleties of language on Twitter can be more easily pinpointed to detect abuse. "What we want to do is be able to identify abuse patterns early and stop this behavior before it starts," said Moody.

IBM Watson uses natural language processing to analyze human speech for meaning, and has the ability to work on huge on repositories of data in a fraction of a second. A partnership with IBM in 2016, which resulted in Insights for Twitter, was the first sign that IBM would begin working on mining the site for sentiment analysis.

The announcement comes on the heels of a scandal facing Google this week, when the company's advertisements were seen in juxtaposition with extremist sites spouting hateful language.

The 3 big takeaways for TechRepublic readers

1. Twitter is partnering with IBM Watson to detect harassment on Twitter before it's been reported.

2. IBM Watson's cognitive computing platform can detect subtleties in language that can alert Twitter when someone is engaging in abusive behavior.

3. Twitter users who are using language that is abusive to bully other users will lose certain privileges on the platform, including having their Twitter feed viewable to only their followers.

Also see...

istock-458594217.jpg
Image: iStockphotos/robas

About Hope Reese

Hope Reese is a Staff Writer for TechRepublic. She covers the intersection of technology and society, examining the people and ideas that transform how we live today.

Editor's Picks

Free Newsletters, In your Inbox