• Wed
  • Dec 24, 2014
  • Updated: 7:36am
LifestyleTechnology

The 'perfect matchmaking formula' for online dating has finally been discovered

A new dating algorithm offers an easier and more scientific way to find that special someone

PUBLISHED : Friday, 06 December, 2013, 8:45pm
UPDATED : Friday, 06 December, 2013, 9:05pm

Looking for love? Researchers at the University of Iowa say that they have found the perfect matchmaking formula.

A new algorithm designed for usage on online dating sites has been developed by a team led by Kang Zhao, assistant professor of management science at the Tippie College of Business.

The algorithm is based on two variables: a user’s tastes and his or her attractiveness. Taste is determined by the type of people he or she previously contacted, while attractiveness is based on previous success rates when initiating contact.

The formula is similar to the model used by Netflix to recommend movies to users based on their viewing history.

According to Zhao, the combination of these criterions will help generate better matches rather than simply relying on information entered into online dating websites.

Online dating profiles can often be intentionally misleading or users may not really know their own preferences in potential partners, he added.

“Your actions reflect your taste and attractiveness in a way that could be more accurate than what you include in your profile," Zhao said in a University of Iowa press release.

For example, a person’s profile may stipulate that they like certain attributes, such as a tall partner, but if the user consistently picks profiles indicating shorter mates, the algorithm will change its recommendations accordingly.

"In our model, users with similar taste and attractiveness will have higher similarity scores," Zhao said. "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."

The algorithm was deduced through an analysis of data provided by a popular commercial online dating company, whose identity is being kept confidential.

47,000 users in 2 US cities, of which 28,000 were men and 19,000 were women, were studied. A total of 475,000 initial contacts were made, with men making the first move 80 per cent of the time.

From the original data, only 25 per cent of initial contact was reciprocated. Using Zhao’s algorithm, the success rate can be increased to 44 per cent.

Of course, science and mathematical deductions can only go so far. In the past, such “matching algorithms” have been criticised.

"There is no compelling evidence that any of these algorithms work," Professor Eli Finkel, a professor of social psychology at Northwestern University, said in 2012.

“Limiting the number of potential partners is only helpful if the algorithmic-selection process favors compatible partners over incompatible ones,” he added.  

Finkel is the lead author of a 2012 study, where he and his team compared online dating with “conventional offline dating” and analysed whether online dating provides better romantic outcomes.

Finkel acknowledged online dating has some benefits, such as an increased pool of candidates. However, according to Finkel’s research, the process merely facilitates the ultimate face-to-face interaction required for partners to gauge romantic potential. In the end, it’s chemistry - not an algorithm - which determines an amorous match.

Still, Zhao’s studies have been well-received, and he has been approached by two online dating companies who are keen to find out more about his algorithm. Zhao and his team are also eager to continue testing the algorithm on different groups of people to refine the system.

Furthermore, Zhao believes that his algorithm can transcend the world of love and has potential applications in the work environment and in university selection processes – where, perhaps, sexual chemistry is less important.

Share

Related topics

For unlimited access to:

SCMP.com SCMP Tablet Edition SCMP Mobile Edition 10-year news archive
 
 

 

 
 
 
 
 

This article is now closed to comments

fanfan.wong1@gmail.com
b

Login

SCMP.com Account

or