It’s unthinkable for a lot of us to live even for a day without social media.
Facebook, Twitter, Instagram, and dozens of other platforms. Personally, I spend hours just randomly browsing whatever comes up in my feed.
It’s well-known that each feed/timeline is algorithmically-tuned for every individual. What I see on my Facebook news-feed can be radically different than a friend’s news feed, even if we have many mutual friends and even if we have overlapping page likes.
It’s simple: Facebook, and many others, show you content, that based on your history, you’re likely to engage or interact with.
Have you been planning a trip and visiting many travel pages on Facebook or websites outside the platform? You’ll find your news-feed filled with posts advertising your next dream-destination (been there myself). Have you been reading a little too much about COVID-19? You’ll find yourself overloaded with news about vaccines, lock-downs, and maybe even about the end of the world as we know it.
Having hundreds of friends or followers is now common across the different networks. Add to that pages and accounts you follow, and you end up with a very large list of people or businesses, who create or share a large amount of content every single day.
Even if we discard ads and suggested content, do you think you’re seeing the entire posts shared by your network? I’m fairly certain that the answer to that is a definite no.
What determines what you see? An algorithm. In the case of recommending posts and determining which ads will be displayed, it is most likely a Machine Learning algorithm.
In simple terms, Machine Learning is when you define a task, data (our browsing history, and tons of others), a metric to optimize (can be maximizing engagement, clicks on ads, etc..), and the performance according to the metric (usually) improves directly the more data you feed it.