In the earlier days, people used to meet in public places and strike up conversations with each other. If they were lucky, they would exchange numbers. Others had to meet each other after their friends or families introduced them to each other. Online dating and matrimonial websites have changed this trend to some extent. Not just that, dating and matrimonial sites have themselves evolved over time to become more data oriented.
For the percentage of people on dating sites, they’re pretty successful in the sense that they are said to find better matches than any person would make in real life, according to the users themselves. Nearly 53% of users, past and present, said that they found better matches online. In mobile platforms, this isn’t exactly the case. Tinder doesn’t have the same success rate as eHarmony, but that could be due to the lower average age of users on Tinder, immature relations may not last as long anyway. Another possible reason could be eHarmony’s detail oriented algorithm. eHarmony never really picked up in India though, something that could be caused by the culture.
How do these websites and apps work?
Tinder works on an undisclosed rating system, similar to chess bots. Users are shown similarly rated people and the rating increases with the number of likes, superlikes and matches that a person gets. This is a straightforward system which is mostly just based on competition rather than giving the users personalized suggestions for people. Dating websites like eHarmony often overcome this flaw by asking for a lot of information and by taking some data, this information is either compared to a user’s information to try to match them or it’s used to filter out certain users. All of the dating websites and apps have a location based approach though. Only people who are supposed to be within a certain distance of each other are allowed to match with each other.
Alternate solutions for matchmaking
There is also an approach of using deep learning in online dating. Apparently, people are attracted to others who look a little bit like themselves. Deep learning can be used to analyze images to find similar faces. This can either be used to find people who look like someone a person happens to like by uploading an image of that person or by finding someone who looks like oneself by uploading one’s own image.
Obviously, the dating and matrimonial apps and websites are trying their best to find “the one” for their users. This doesn’t mean that people won’t stop lying about their height or weight in an attempt to try to seem more appealing. Lack of verification is probably the biggest problem with this but it should improve as people start getting more confident about themselves (hopefully that’ll happen someday).
(The article is authored by Nishant Nijhawan, Intern, Techxty.com. Nishant is a student of computer engineering at Delhi Technical University who believes in optimal usage of technology.)