Each day brings fresh confirmation that giving away any information means you’re giving all of it away, thanks to algorithms that crunch your language use, purchasing habits, friend structure, and other bits of information.
Now “big data” can see inside your heart.
The latest offering from Facebook’s data-science team (see “What Facebook Knows”) teases out who is romantically involved with whom by examining link structures. It turns out that if one of your Facebook friends—let’s call him Joe—has mutual friends that touch disparate areas of your life, and those mutual friends are themselves not extensively connected, it’s a strong clue that Joe is either your romantic partner or one of your closest personal friends. Details can be found here.
The company says this could help them decide which posts to give extra prominence in news feeds. “If we can do a better job of identifying all the most important people in your life, there is a lot of opportunity to make Facebook better,” says Lars Backstrom, a data scientist at Facebook.
Surely, it will also be of value to Facebook’s advertisers.
The finding adds to the pile of other evidence of data-mining possibilities on Facebook. A study this year by University of Cambridge researchers, using data from 58,000 volunteer U.S. Facebook users, found it could predict traits like race, sexuality, substance abuse, and parental separate, using Facebook “likes.” And others are teasing out demographic cues such as gender and age from the kind of language you use (see “How Your Facebook Profile Reveals More About Your Personality Than You Know”).