Facebook and its famed News Feed Algorithm
In 2006, Facebook built the news feed as a hub for updates about your friends’ activities on the site. But then another problem arose, that of overwhelming people with hundreds of updates every day. To counter this, Facebook came up with a crude algorithm to filter them based on how likely they were to be of interest. However, there was no real way to measure that and the like button came three years later. The like button wasn’t just a way for users to interact on the site, it was a way for Facebook to enlist its users in solving the problem of how best to filter their own news feeds.
The like button gave Facebook a way to identify the most popular posts and make them go “viral”. However, Publishers, advertisers, hoaxsters and individual users began to know realize the elements that made a post go viral. As such, many began to tailor their posts to get as many likes as possible. It wasn’t long before Facebook users’ feeds began to feel eerily similar: all filled with content that was engineered to go viral. The substance, nuance, sadness and anything that provoked thought or emotions beyond a simple thumbs up began to get drowned.
While engagement metrics were up, the news feed wasn’t optimizing for the metrics that actually mattered to Facebook’s ever growing user base. Something had to be done. Over the next ten years, Facebook worked to fine tune its news feed algorithm. The result? An algorithm that is way better and whose optimizations are more tailored to what each user wants to see on their news feeds.
The algorithm was made to take in more “variables” or features as they are called in machine-learning lingo. Hundreds in fact. Unlike Google’s PageRank which primarily calculates the page rank based on the number and quality of links leading to a particular page, Facebook’s news feed algorithm has to take more into account. This is because it is dealing with humans on the other end and not just pages. What if people “like” posts that they don’t really like, or click on stories that turn out to be unsatisfying? The result could be a new feed that optimizes on “virality”, rather than quality.
To make sure the algorithm meets the specifications set out, it had to be robust and yet flexible. The algorithm doesn’t just predict whether you’ll actually hit the like button on a post based on your past behavior. It also predicts whether you’ll click, comment, share, hide or even mark it as spam. It will predict all these outcomes and others with a certain degree of confidence, then combine them all to produce a single relevancy score that’s specific to both you and your post. This is where the similarity to Google’s page rank algorithm comes in. Once every possible post in your feed has received its relevancy score, the sorting algorithm can put them in the order that you’ll see them on the screen. The post you see at the top of your feed has chosen over thousands of others as the one most likely to make you laugh, cry, smile, click, like share or comment.
Source: http://www.slate.com/blogs/future_tense/2016/10/26/facebook_s_revamped_safety_measures_aim_to_prevent_suicides.html