Article: Can Algorithms Really Master Matters of the Heart?

Photo Source: Negative Space

Singles looking for love across the nation today depend on the massive computing power of the many online dating applications that have emerged in the past twenty years. Online dating ranks as the most popular way to meet people in the US, surpassing the old-fashioned way of going through friends and family. The variety of dating sites continues to expand from the well-established eHarmony and Match to Tinder, Bumble, and Hinge. To put this growth in perspective, in 2018, 40 million Americans have tried an online dating site, according to the Pew Research Center. In 2022, Bumble alone has 45 million active users, 1.5 million of which subscribe to the app’s premium features (businessofapps.com). The most popular dating app, Tinder, has over 70 million active users in the US alone! eHarmony claims responsibility for 4% of new marriages in the US annually. (businesswire.com)

Although similar in some ways, different dating apps will have their own algorithms to help users find suitable matches out of the millions of possible dates. The term algorithm gets tossed around a lot, but it is just a fancy term for the steps needed to make something. For example, the algorithm in a bread-making recipe consists of the steps of mixing flour, yeast, and salt followed by kneading, proofing, and baking required to produce a loaf of bread. Different types of bread have slightly different ingredients and algorithms. Similarly, in dating applications, algorithms provide the machine with the proper steps for processing information such as age, preferences, and interests to suggest the best matches.

The online dating application eHarmony has served millions of customers since its launch in 2000. eHarmony’s algorithm uses mathematics to match couples with similar qualities, beliefs, and interests. The algorithm relies on research into the qualities of 5,000 happy, successful couples. As a result, the company believes that similar couples succeed. The vice president of Research & Development at eHarmony said in an ABC interview, “Opposites attract, then they attack. Over time, if you have differences you’re going to constantly have to negotiate them. Those differences start to be not so cute.” eHarmony gets its information from customers filling out an extensive questionnaire and by observing how customers use their application. (abcnews.go.com) eHarmony recommends possible matches but does not allow members to browse through profiles.

The second most popular dating app in America, Bumble, takes a very different approach than eHarmony. Bumble allows users to browse through profiles with photos of its users. If someone looks interesting, the user will swipe right or swipe left if not interested. The Bumble algorithm pays attention to how often and for how long someone uses the app and who you swipe right or left. This information allows the algorithm to rank the likelihood of a match. This algorithm, called the Elo algorithm, also gets used for ranking chess players to predict who will most likely win in a series of games. According to an article titled “Understanding the Bumble Algorithm in 2022 to Get Results,” the author, Johann Davis, determined that this application caters to an older clientele looking for serious relationships. (beyondages.com) He notes that posting the maximum number of photos allowed and maximizing time on the app will get the best results.

The most popular dating app, Tinder, no longer uses the Elo ranking system used by Bumble and in chess player ranking, according to their blog post. (tinderpressroom.com) Tinder’s algorithm looks at several things, including interests added by the users to their profiles. Also, Tinder will analyze photos for similarity to ones a user has liked. For example, if a user liked photos taken at the beach or in the mountains, the algorithm will recommend people in similar settings. Additionally, Tinder pays close attention to how much time someone spends on the app. They prioritize active users of the app and use proximity to drive recommendations.

The dating app has now surpassed the traditional way of meeting people—through friends and family. Tens of millions of people use a growing array of dating applications to find love and friendship in America. Traditional apps such as eHarmony match people based on detailed questionnaires and the philosophy that similar people make better matches. Newer apps such as Bumble use an algorithm to rank the likelihood of a match based on who a user has swiped right to like or left to not like. The most popular dating app in the US, Tinder, looks at a complex mix of likes and not-likes, profile preferences, the contents of posted photos, and proximity. Different apps use different algorithms, and doing some homework to see what drives them can make for more successful dating. They all say that the more time you spend on the app, the better the recommendation.

Dr. Smith’s career in scientific and information research spans the areas of bioinformatics, artificial intelligence, toxicology, and chemistry. He has published a number of peer-reviewed scientific papers. He has worked over the past seventeen years developing advanced analytics, machine learning, and knowledge management tools to enable research and support high-level decision making. Tim completed his Ph.D. in Toxicology at Cornell University and a Bachelor of Science in chemistry from the University of Washington.

You can buy his book on Amazon in paperback and in kindle format here.