Cornell University researchers have developed algorithms that can correctly identify a person’s spouse, fiance, or other romantic partner, based on a map of Facebook friends, with about 70-percent accuracy. “We are trying to build up a sort of chemistry kit for finding different elements of a network,” says Cornell professor Jon Kleinberg. He says the method works best when the couple is married, and works better the longer the relationship has been in force. However, if the algorithm does not select the person who is the relationship partner, there is a significantly increased chance that in a month or two the couple will break up, according to the researchers. The researchers tested their algorithms on anonymized data from 1.3 million randomly selected Facebook users aged 20 or older who listed their status as “married,” “engaged,” or “in a relationship.” The algorithms combine embeddedness and dispersion, as well as the dispersion of mutual friends, to determine if two people are in a relationship. The team will present their research at the ACM Conference on Computer Supported Cooperative Work and Social Computing taking place this week in Baltimore.
More info: Cornell Chronicle (02/12/14) Bill Steele