Facebook Updates News Feed Algorithm to Eradicate Clickbait Headlines

Datetime:2016-08-23 03:14:56          Topic: Algorithm           Share

Translate

Facebook has updated its algorithm to eliminate clickbait headlines from its News Feed.

You know the ones— those sensational headlines that lead to inaccurate and misleading content once clicked on.

The update will also push relevant posts, like those from friends or pages users like, higher up in the feed results.

“We are updating News Feed by using a system that identifies phrases that are commonly used in clickbait headlines,” writes Facebook research scientist Alex Peysakhovich and user experience researcher Kristin Hendrix in a blog post.

If a headline withholds information required to understand what the content of the article is or exaggerates the article to create false expectations for the reader, then Facebook will categorize it as clickbait.

“For example, the headline ‘You’ll Never Believe Who Tripped and Fell on the Red Carpet…’ withholds information required to understand the article (What happened? Who Tripped?) The headline ‘Apples Are Actually Bad For You?!’ misleads the reader (apples are only bad for you if you eat too many every day),” the blog post reads. “A team at Facebook reviewed thousands of headlines using these criteria, validating each other’s work to identify a large set of clickbait headlines.”

Next, Facebook uses its newly-built system to look at the set of clickbait headlines to determine what phrases are commonly used in clickbait headlines that are not used in other headlines.

Once the system identifies posts that are clickbait and which Web domains and Pages these posts come from, all links posted from or shared from these Pages or domains will appear lower in the News Feed.

If a Page stops posting clickbait headlines, however, its posts will stop being identified as clickbait.

Facebook said most Pages won’t be impacted as a result of the algorithm change — only those that rely on clickbait-style headlines will see a distribution to decrease.





About List