Facebook knows its members very well. And Facebook is trying to bring them together. Under the heading “People you may know”, each member receives regular suggestions for friend requests – now also on Instagram.

Work colleagues appear in the list next to party acquaintances or old school joy. In between, however, there are always completely unknown people – the choice of people is surprising at first glance. And we don’t even want to be friends with many people suggested to us. But how does Facebook know that we know these people? And how are they selected?

This is how Facebook gets the data

Facebook pulls its data from various sources in order to create individual suggestions. The results are calculated using a complex algorithm, of which little has so far been made public. One goal of the “People You May Know” section is to increase Facebook advertising revenue through more networking, more interaction and longer visit times.

It is not known exactly how the algorithm works. On Facebook, however, the following is considered safe: mutual friends, the overall structure of the previous network of friends, data on training and work, group memberships and the contacts that you (possibly unconsciously) imported from your smartphone address book to Facebook. “

An absurd example of friendship proposals is known from the USA. Facebook wanted to connect the patients of a psychiatrist with each other, even though they didn’t know each other. Even with the psychiatrist as a possible link, they had no contact via the social network, according to a “Fusion” report. One explanation for this would be that several patients had the doctor’s number saved in their cell phone – and allowed Facebook to access the smartphone contacts.

How exactly does the data comparison work?

For example, Facebook compares the voluntary profile information with each other. Were members in the same school? Do you work in the same company? Did you grow up in the same place? If this information has not been provided directly, the software can also create a profile for the user through a few friend requests and, for example, suggest other work colleagues.

Facebook members give a lot of data voluntarily, because they gave the company the permissions for it when they registered. This includes, for example, personal telephone numbers and e-mail addresses. Facebook also uses imported contacts from the smartphone’s address book.

Another factor is the number of mutual friends. If two people have several common acquaintances, then they should connect as well, says Facebook. For invitations and events that the person attends, Facebook also connects to other participants and suggests them as new friends.

Link Prediction finds friends with whom there is no direct contact

To find friends with whom there is no direct connection, the Facebook algorithm uses the mathematical method Link Prediction. Every “I like” entry, every status message and every search entry is included in the calculation. For this, Facebook even uses everything a member enters in the search bar. EVERY interaction within Facebook is decisive for Link Prediciton – even if it was only typed in and never sent: Since the Javascript code transmits all characters entered immediately, Facebook ALWAYS knows immediately – even if I repeat my entry in search or chat delete before I send it off.

Through all of these measures, Facebook can specifically suggest acquaintances as friends that users have met outside of the network. Even people about whom it actually cannot know anything. Sometimes it is enough if two people are in the same place to have the other as a suggestion in the timeline.