clock menu more-arrow no yes

Filed under:

Hinge’s newest feature claims to use machine learning to find your best match

New, 6 comments
Image: Hinge

Hinge’s newest feature — Most Compatible — attempts to use all your cumulative data to find the perfect match for you. The company’s been testing this feature, which occasionally recommends a possible match to users, for at least month now. Those recommendations were only offered once a week during testing but will now come every day. Justin McLeod, Hinge’s CEO, tells me the company spent the testing time honing its backend algorithm and getting Most Compatible to a point where the company feels confident putting it fully out there.

Most Compatible, he says, uses machine learning to figure out each user’s taste. It primarily relies on the Gale-Shapley algorithm to determine every user’s perfect match. The idea is that, essentially, there is a user out there who is most likely to like you and you’re most likely to also like. That would make you a good match in Hinge’s world. The company’s technology breaks people down based on who has liked them. It then tries to find patterns in those likes. If people like one person, then they might like another based on who other users also liked once they liked this specific person. Both users will receive the same recommendation on the same day, and it’ll expire after 24 hours.

The email Hinge will be sending to users about the new feature.
Image: Hinge

I use Hinge and have tested this feature for the past month or two. It’s kind of exciting when you get a recommendation, but at the same time, I never understood where it was coming from. I wish Hinge could explain why I would make a good match with this person, but in reality, the tech just probably knows I’ll find them attractive and can’t really go off much else. It doesn’t know how our conversation would go or whether we have mutual interests. OkCupid, for example, asks users to answer lots of questions about themselves and then determines users’ compatibility based off their responses. Hinge doesn’t have that kind of information. Instead, every users answers three questions about themselves and uploads six photos. Maybe in the past I’ve liked some user’s witty answers, but Hinge isn’t breaking down my likes in that way. It just knows that I like this person and that I might like other similar people. I will say that most of the recommendations I receive do seem on point, but I haven’t gone out with anyone that Hinge suggested.

As much as I appreciate a machine sifting through the New York City masses to find a person for me to date, I also feel conflicted about the fact that I’m so predictable. Is it really that easy to figure out who I will find attractive based on my previous likes? I guess so. I wish I was more complicated.

Correction 7/11, 3:16 PM ET: This article previously stated that recommendations didn’t expire, but that was incorrect. Users have 24 hours to act on a possible match before it disappears.