Contributed by: ElyseRomano on Friday, August 29 2014 @ 06:52 am
Last modified on Wednesday, May 24 2023 @ 12:13 pm
All of us, at least once, no matter how much we profess to love dating, have wished we could just snap our fingers and magically make the perfect partner appear.
Well…I have good news and I have bad news. The good news is that researchers at the University of Iowa believe they may have found the secret to online dating success. The bad news is that no magic is involved, and snapping your fingers probably won’t help.
The key, says the University of Iowa team, is for online dating sites to match users based on their past interests and actions, rather than on what they say they’re interested in. Think about your Netflix account. Netflix recommends movies based on what you’ve watched and the amount of time you spent watching it – so even if you profess to be a documentary lover, if all you watch is Gossip Girl, Netflix is probably going to suggest more CW-style teen dramedies.
The online dating algorithm developed by Kang Zhao, an assistant professor at the Tippie College of Business, and doctoral student Xi Wang works in a similar way, using a person’s contact history to recommend future dates. Together they looked at 475,000 initial contacts involving 47,000 American users over 196 days from a popular online dating site. The data showed that only about 25% of those initial contacts (80% of which were initiated by men) were reciprocated.
Determined to improve that rate, the researchers developed a new model that combines two factors to recommend dates:
Together, says Zhao, the combination of taste and attractiveness is a better predictor of connection and compatibility than the information a user enters into his/her profile. Turns out, we’re not so good at that “know thyself” thing. You may think you like tall partners, but in reality you may always gravitate to vertically-challenged dates. It could be that we lie to ourselves about what we like, but it could also simply be that we know ourselves less well than we think. Either way, this is a clear case of actions speaking louder than words.
“In our model, users with similar taste and attractiveness will have higher similarity scores than those who only share common taste or attractiveness,” Zhao says. “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.”