Making a Connection: Pairing People with Similar Interests in Online Dating World

Thanks to researchers from the University of Iowa, they've stumbled onto a "matchmaking formula." It's really quite simple: Pair people according to past interests in the online dating world instead of who they think they're interested in.

Lead study author Kang Zhao, assistant professor of management sciences in the Tippie College of Business, and UI doctoral student Xi Wang, are part of a team that worked to develop an algorithm for dating sites that used a person's contact history in order to recommend partners with whom they may be more compatible.

In fact, the equation allegedly works so well that they've already been contacted by two dating services that are interested in learning more about the model.

Researchers looked at 475,000 initial contacts involving 47,000 users in two U.S. cities over a 196-day span. Of the users, 28,000 were men and 19,000 were women, and men made up 80 percent of the initial contacts.

Zhao believes that the data consists of about only 25 percent of the initial contacts that were actually reciprocated. In order to improve the rate, Zhao's team developed a model that combines two factors to recommend contacts: a client's tastes, determined by the types of people the client has contacted; and attractiveness/unattractiveness, determined by how many of those contacts are returned and how many are not.

"Your actions reflect your taste and attractiveness in a way that could be more accurate than what you include in your profile," Zhao said, via a press release. "In our model, users with similar tastes and (un)attractiveness will have higher similarity scores than those who only share common taste or attractiveness."

He adds that "the model also considers the match of both taste and attractiveness when recommending dating partners. Those who match both a service user's taste and attractiveness are more likely to be recommended than those who may only ignite unilateral interests."

Research data showed a return rate of 25 percent. Zhao said he believes that a recommender model could also improve the returns to up to 44 percent.

More information regarding the study can be found via the journal IEEE Intelligent Systems.