This Matchmaker App Uses AI to Expedite Love

By Didem Tali, Contributor

Like many single professionals in the San Francisco Bay Area, Bahar Mir looked to dating apps to meet new people and find romance. Yet, it didn’t take too many swipes for Mir to find herself tied to an app that wasn’t producing the results she wanted.

“You spend lots of time messaging back and forth and usually nothing happens,” the engineer-turned-entrepreneur said. “Especially for busy professionals who don’t have a lot of time in the first place, this can be very disappointing.”

A former product lead at Google, Mir realized how many of her friends had similar frustrations, so she set out with a new goal: She wanted to use artificial intelligence (AI) to eliminate the real-life bugs of dating, such as endless messaging and flaky connections.

In August 2017, Mir launched her dating app—BELONG—one of the first to use AI to make more relevant matches.

With users’ permission, BELONG’s algorithms analyze their demographic and behavioral data. From photos and social profiles to online interactions and event attendance, BELONG sets out to provide more relevant recommendations and (in theory) expedite introduction to an ideal match.

BELONG’s functionality comes from Mir’s research during the production stage which, she said, took her about a year. During this phase, she found that 42 percent of online daters felt an “overload of unfit choices” each time they logged into the app.

What’s more, 54 percent of online daters were frustrated with “flaky users” that weren’t serious about meeting in real life, or creating a meaningful relationship. One of the biggest challenges of online dating, she explained, was getting the first date to actually happen.

“People just get busy and lose interest somehow,” she said. “A lot of people also feel awkward about arranging a date. They don’t know what to do.” To crack the issue of getting people to meet face-to-face, Mir looked to machine learning to make it easier and less awkward for people to get themselves out of the house and onto that first date.

The more information the user provides, the smarter the app gets about providing match recommendations. Beyond identifying matches, the app acts as a concierge that provides smart recommendations about activities—the data indicates—both parties might enjoy. As the matches message each other, they receive personalized recommendations about events, such as concerts, cooking, or a yoga class they could do together.

Development of this app feature was central to Mir’s mission to get people on that first date and expand their lives as a whole.

“Everyone needs to connect in real life. Our plan is to get smarter with all aspects of social life.”
— Bahar Mir, BELONG Founder

But what about after that first date? The app asks users for feedback post-date to be able to better mimic positive experiences and mitigate for those that are subpar, or correct for flakiness.

“Whether the users got along or not is a private matter,” Mir mused. “But if one of them didn’t turn up or ghosted the other one, our system learns about it.”

Mir doesn’t promise users will find the love of their lives on BELONG. Instead, she offers an intelligent approach to dating that aims to replace mystery with data and provides a romantic concierge.

“Everyone needs to connect in real life,” Mir said. “Our plan is to get smarter with all aspects of social life.”