X
Business

Facebook wants to rid of engagement bait with machine learning

Facebook wants people to stop posting content that encourages users to interact with likes, shares and comments.
Written by Jake Smith, Contributor
facebook-fighting-engagement-bait.png
Image: Facebook

Facebook said Monday it's working to rid of "engagement bait" posts -- that ask for users to interact with likes, shares, and comments -- with machine learning.

Teams at Facebook have categorized "hundreds of thousands of posts" to inform its machine learning model to detect different types of engagement bait. Facebook said posts that use the baiting tactic will be shown less in users' news feed.

In the coming weeks, Facebook also plans to implement stricter demotions for Facebook Pages that repeatedly use engagement bait in their posts.

"Publishers and other businesses that use engagement bait tactics in their posts should expect their reach on these posts to decrease," Facebook warned in a blog post.

The removal of engagement bait comes amid discussion about the social network's impact on a person's health and well being.

Last week, Facebook published academic research on social media and mobile phone impact in people's lives. Facebook said its ultimate goal has always been to be a place for meaningful interactions with your friends and family. Further, it shared insights into how the research team at Facebook works with our product teams to incorporate well-being principles.

In recent months, Facebook has also made efforts to show less clickbait headlines and links to low-quality web browsing experiences. Facebook CEO and founder Mark Zuckerberg said in November 2016 that his company would work to combat the spread of misinformation, which critics claim spread rampant during the US presidential election.

"The bottom line is: We take misinformation seriously," Facebook's chief explained last year. "Our goal is to connect people with the stories they find most meaningful, and we know people want accurate information. We've been working on this problem for a long time and we take this responsibility seriously. We've made significant progress, but there is more work to be done."

Previous and related coverage

Twitter launches feature to organize multi-tweet threads

You can now string together your "tweetstorms" with Twitter's new threads features.

Facebook's AR Studio is now open to all third-party developers

AR Studio was introduced in April at Facebook's F8 conference as part of the Camera Effects platform.

Editorial standards