Constantly plagued with privacy issues, Facebook might actually hold secrets that romantics might want to know. The social network can play the role of a good fortune teller for those wondering who a good match might be for them and when that person will dump them. This is implied by a new study conducted by experts from Cornell University and Facebook.
Using Facebook, the researchers could accurately predict who are couples, based not on the number of mutual friends, but on dispersion of social ties that looked into the quality of networks the two people share. Using the algorithm they developed, the proponents of the study were able to identify the spouse of a Facebook user 60 percent of the time while those who declared that they are in a relationship can be pinpointed accurately about 33 percent of the time.
A fascinating part of the results is when the dispersion algorithm fails. This seemed to indicate that a relationship might also be problematic. A couple who declared a relationship but without high dispersion on Facebook has a 50 percent chance to part ways within two months compared to couples with high dispersion. The results also indicate that couples who build an integrated life tend to last longer.
The study titled "A Network Analysis of Relationship Status on Facebook" published Sunday, involved 1.3 million users of Facebook who were randomly selected from users with a minimum age of 20, with friends number between 50 and 2,000, and who have profiles indicating that they are in a relationship. The tandem of Lars Backstrom, a News Feed engineer at Facebook, and Jon Kleinberg, a computer expert from Cornell University, conducted the study and analyzed around 8.6 billion connections and over 379 million information nodes. They used the data anonymously.
"A spouse or romantic partner is a bridge between a person's different social worlds," said Kleinberg in an interview.
The research may help Facebook tailor content and advertisements for users; however, the social network is not yet using the information gathered from the study.
"The specific findings from this study are not currently being used in making News Feed a better product. It has helped the team think about things and informed some of the features we use in ranking, but we are not using the specifics in the paper," a Facebook spokesperson said in a statement.
The paper will be presented during a social computing conference scheduled in February 2014.
