Scientists may predict how your romantic entanglements on Facebook will turn out

Posted by Emory Kale
A new research study - jointly written by Lars Backstrom of Facebook and Jon Kleinberg of Cornell University - tries to understand how your relationships pan out figuratively and mathematically. The result is a predictive of the ebb and flow of your relationships.
 
A social network is a set of links and connections between you and your network of family, friends and business associates. The more you use online networks to manage your links to your networks, the easier it is to discern patterns of engagement and disengagement. Romantic Partnerships and the Dispersion of Social Ties: A Network Analysis of Relationship Status on Facebook by Backstrom and Kleinberg looks at the complexity of the relationships in your networks and tries to find the forces at play.
 
Backstrom and Kleinberg decided to focus on romantic partners to identify how the roles of significant people in your online social network neighborhoods define the structure of your relationships. The measure they came up with is dispersion, according to their paper.
 
Dispersion is a measure of how engaged nodes in your social network can be at any given point in time. It applies to everyone who comes in and our of your online social life. People who don't necessarily fit in easily within your network, and who are not easy to classify within your network, have a high dispersion value. This is because these people may span across many contexts of your social life, either because they were around at many points in your life, or because they are included in multiple social circles. So, romantic partners and spouses have high dispersion values, as do family members.
 
Seeing as this is Facebook research, this ain't no altruistic science paper. Dispersion has application in the real world. It may help identify ways to identify how to annotate users with a high dispersion value to determine what really engages them within your network. It raises the efficacy of targeting content, and most likely, ads.
 
 
The research also identifies how the people closer to us cannot be easily clustered into any grouping around us. They defy easy categorization within our social networking groups because, we include them and engage with them on many different levels. There are bridges and impediments to traversing these clusters for everyone, meaning that just because your buddy likes to watch Football on a Sunday with you, he is not going to be too happy about spending the evening with you and your girlfriend watching Scandal. 
 
Understanding the way the nodes of our networks, our friends, family, associates and partners interact, or not, is useful in predicting who we are truly connected to hence, a drifting girlfriend who has her own nodes to contend with, may be on her way to checking out of a relationship.
 
Instinctively, we know when someone has lost interest. They hang out with other people. The things you found fun and interesting about each other become annoying and irritating. 
 
It's very likely, according to this research, that we can see the pattern of these shifts in emotion by understanding how each person's network is functioning.
 
According to the New York Times:
 
Their dispersion algorithm was able to correctly identify a user’s spouse 60 percent of the time, or better than a 1-in-2 chance. Since everyone in the sample had at least 50 friends, merely guessing would have at best produced a 1 in 50 chance. The algorithm also did pretty well with people who declare themselves to be “in a relationship,” correctly identifying them a third of the time — a 1 in 3 chance compared with the 1 in 50 for guesswork.
 
Particularly intriguing is that when the algorithm fails, it looks as if the relationship is in trouble. A couple in a declared relationship and without a high dispersion on the site are 50 percent more likely to break up over the next two months than a couple with a high dispersion, the researchers found. (Their research tracked the users every two months for two years.)
 
Yay, Facebook. It's not exactly that sound a prediction mechanism, just a tad over 50/50, but I could use the help. Just let me know why the toilet seat is up every Wednesday when I get home from work. I know it isn't me.