Sitemap
Data Science Collective

Advice, insights, and ideas from the Medium data science community

Member-only story

The Friend Suggestion That Explains Graphs Theory Better Than books

16 min read2 days ago

--

In 2016, a reporter told the story of a psychiatrist who opened Facebook and found her own patients in her “People You May Know” box. Then it got stranger, because Facebook began suggesting the patients to each other. They had never met. They’d only sat in the same waiting room on different days, and the likeliest link was one phone number saved in all their contacts.

Screenshot of Facebook People You May Know friend suggestions, a real recommendation feature built on a social network graph
The People You May Know box, the friendly little feature that once matched a psychiatrist's patients with each other

Free reading link here

So nobody at Facebook read a medical file. A phone book was enough.

Every patient had one contact in common, the psychiatrist, and to a computer a shared contact is all it takes to make two strangers look connected. By the end of this article you’ll know the exact calculation behind that, and it’s a single small multiplication.

The thing doing the work is called a graph. Your phone is full of them, so let’s open a few apps and look inside.

Press enter or click to view image in full size
Cartoon robot drawing glowing lines between startled strangers in a waiting room using their phones, a social graph recommendation
An over-keen matchmaking app connecting everyone who has ever shared a waiting room, a phone number, or a bad haircut

Dots, lines, and your group chat

--

--

Kamrun Nahar
Kamrun Nahar

Written by Kamrun Nahar

I write about machine learning, deep learning, and statistics with Python, making complex ideas ridiculously clear.

Data Science Collective
Data Science Collective

Published in Data Science Collective

Advice, insights, and ideas from the Medium data science community