Dating vs Data - Can an Algorithm Help you Find True Love?

You spend hours swiping left and right on Tinder or Bumble on your phone. You go on date after date only to find that the initial chemistry you felt from a few profile photographs and text flirtation fades rapidly once you meet someone in person. You're annoyed, yet you keep swiping or scrolling in the hopes that technology will solve all of your marital woes. Dating apps, like many contemporary technological advances such as Uber, Facebook, and Mint, use algorithms to make life easier—in this case, navigating the dating minefield to find your love. So, why is it so much more difficult to create an algorithm to help you find love than it is to construct one to help you find a ride? Chemistry is the short answer.

Understanding Chemistry

Source

Our physiology governs the chemistry that governs our attraction to one another. Humans are more attracted to mates who have distinct immunity genes, according to studies (so that hypothetical children would have stronger immune systems). According to other studies, women are drawn to males with high testosterone levels, whereas men are drawn to women with high copulin levels. So, if a date isn't going well and nothing seems to be clicking, don't take it personally; it's subconscious. When sparks fly, the body goes through two different chemical reactions. The heart begins to beat quicker, the liver releases glucose, and the brain makes a split-second "fight or flight" decision in a promising first-time encounter. The senses are involved in the second chemical reaction, which might happen soon after the first. Each sense looks for specific traits and votes "Yes" or "No" based on them. Our eyes, for example, look for signs of excellent health, our ears listen for a certain voice pitch that is acceptable, and our nostrils look for remnants of certain odors. The chances of a chemical match decrease if any of the senses vote "No."

Can Dating Apps Predict Chemistry?

Source

Is it viable for dating apps to mimic even a human matchmaker's attempt at chemistry prediction? It is, according to Sebastien Koubar. He's the co-founder and CEO of Meetwo, a dating app that claims to "transform the way people connect online by leveraging chemistry," according to its slogan. "It's not an algorithm." Love is a difficult concept to define and comprehend, so how could a machine accomplish it? Meetwo is attempting to address this issue by asking people who sign up for an account a random set of yes-or-no questions about themselves in order to evoke genuine emotional responses. "Do you think women should have to do the dishes?" will be one of the questions to "Do you think the man should always pay the bill on a date?" and "Should children be given a smartphone at a young age?"?

Going Beyond “Swipe Right”

Source

Machine learning algorithms learn from in-app user behavior data, which is augmented with profile data, to give increasingly relevant and curated matches. The complexity of such pairing has gone well beyond "swiping right" for attractive people in adjacent areas. Many businesses fine-tune their algorithms to take into consideration personality traits, music preferences, beliefs, attitude, and other factors. Coffee Meets Bagel matches people with a "friend-of-a-friend" match. The company's matching technology utilizes a "blended" strategy and is powered by a deep neural network. The matches are rated by nine models, and the system runs them all and returns a converged score. Men can choose from up to 21 matches (or "bagels") per day, whereas women only get four.

Taking Dating Offline

Hinge wants to be known as a service that helps you get off dating apps as soon as possible. Its Most Compatible function uses machine learning and the Gale-Shapley algorithm to offer daily recommendations for people based on their previous app behavior. In 2018, Hinge introduced a new feature called "We Met," which required matched users to complete a brief private survey about whether they had met offline. This step acts as a feedback loop, teaching the matching algorithms what constitutes effective offline matches over time. After conversing online for a while, eHarmony plans to develop an AI-enabled tool that encourages users to suggest meeting in person. Loveflutter, on the other hand, proposes first date locations that are equidistant from both people's residences.

Leveraging NLP

Source

While some apps refuse to read messages, others do so in order to provide better matches and a better experience. Crushh and Mei Messaging Apps use artificial intelligence to monitor messaging interactions and patterns in order to provide practical insights to users all over the world. Loveflutter intends to analyze user discussions in order to assess their compatibility and suggest a time for them to meet. AIMM takes it to the next level. This is the first fully conversational AI-based app, with the most up-to-date facial recognition and a conversational design. To find your ideal match, AIMM asks you a series of questions and listens to your voice. The software will then begin introducing matches one by one, and will assist you in scheduling a phone call with your match. It guides users through their initial phone contact, offers dating guidance, and even provides feedback later.