Facebook-based dating app Hinge will use machine learning for love

人の出会いは魅力的なものです。初めてでまだ相手をよく知らないほど、相手に惹かれていくものです。浮気は出会いの関数です。夫や妻、彼氏や彼女、どんな人にだって浮気の機会はあるでしょう。もしパートナーの怪しい言動が目に付いたら浮気調査を探偵に依頼してみましょう。その半数以上は無料相談だけで解決しています。
The Most Compatible feature factors how people act on the app (such as who you previously liked) and wants to serve as a virtual matchmaker.

Image: golden sikorka/shutterstock

Hinge, the Facebook friend-based dating app, has been testing machine learning to make better matches for singles.

The feature is called Most Compatible and according to multiple reports, plans to use all your data (there’s a whole trove of it on Facebook) to match people with each other. Most Compatible has been tested once a week for at least this past month, but it will now become a daily feature.

Hinge founder Justin McLeod said that this new feature mainly relies on the classic item matching algorithm Gale-Shapley, which was developed in 1962 and is nickname the stable marriage algorithm. It basically tries making successful matches by choosing the most seemingly compatible person.

Although that algorithm and methodology of using all cumulative data to create matches seems smart — and almost even obvious — the gradual change of Facebook’s role in your life could lead to spotty results. People rarely update their likes now and mainly use it as a tool for keeping tabs on others or curating a social media presence.

How successful would a match over Justin Bieber and One Direction be, when one person liked those pages in high school, while the other un-ironically sings Baby and What Makes You Beautiful in the shower? What about Sports on Facebook, which many people were forced to like in order to access some dumb quiz three years ago?

The Most Compatible feature does, however, take into account how people act on the app (such as who you previously liked). It wants to serve as a virtual matchmaker and aims to find people similar to those you previously matched with on the platform.

Who knows whether this will work. Hinge claims that although it likes having many users, it hopes people get off the app as soon as possible (assuming you leave because your needs have been met). Maybe AI and machines are the friends you never knew you needed to make that first introduction or connection happen.

Https%3a%2f%2fvdist.aws.mashable.com%2fcms%2f2018%2f5%2fd85259b4 dabb d7d6%2fthumb%2f00001