Algorithmic recommendation systems on social media sites such as YouTube, Facebook, and Twitter are responsible for spreading misinformation, propaganda, conspiracy theories, and other harmful content. Facebook, in particular, has come under criticism in recent days for allowing the QAnon conspiracy groups to flourish on its platform. Today, the social network is trying to combat claims that its recommendation systems are somehow to blame for the way people encounter disturbing, unwanted, dangerous, and misleading content. For the first time, the company announced how content recommendations work, TechCrunch reports.
New documentation available from the Facebook Help Center and the Instagram Help Center outlines how Facebook and Instagram algorithms work to filter content, pages, groups, and events for user recommendations.
Suggestions may appear as pages you might like, “Recommended for you” posts in your news feed, people you might know, or groups you should join. Instagram suggestions can be found on Instagram Explore, Accounts You May Like, and IGTV Discover.
The company says Facebook’s existing guidelines have been in effect since 2016 as part of a strategy it calls “remove, cut and inform.” This strategy focuses on removing content that violates Facebook’s community standards, reducing the spread of problematic content that does not violate its standards, and providing people with more information so they can choose what to click, read or share, Facebook explains.
Facebook’s new documentation details five key categories that can’t be included in recommendations. Instagram’s rules are similar. However, the documentation doesn’t provide a deep understanding of how Facebook actually chooses what to recommend to a particular user. This is a key point in understanding recommendation technology, and one Facebook deliberately left out.
One obvious category of content that cannot be recommended includes content that interferes with Facebook’s “ability to create a safe community”, such as self-harm, eating disorders, violence, and the like, and sexually explicit content. and regulated content such as tobacco, as well as content used by deprecated accounts or organizations.
Facebook also claims to discourage sensitive or low-quality content, content that users often dislike, and content associated with low-quality posts. These additional categories include things like clickbaits, deceptive business models, payday loans, products with exaggerated health claims, or offering “miracle cures”, content promoting beauty treatments, contests, giveaways.
Additionally, Facebook says it will not recommend fake or misleading content, such as vaccine disinformation and content that encourages the use of fake documents.
It says it will also “try” not to recommend accounts or organizations that have recently violated community standards.
Given Facebook’s track record, it’s worth considering how well Facebook can follow its own set of rules. It is also worth noting that there are many gray areas that these guidelines do not cover.
Conspiracy groups are just a couple of examples.
For example, a quick search for “vaccines” now suggests a number of groups focused on alternative treatments and anti-vaccine groups.