A few months after an engineer created a robot whose sole purpose of existence is to swipe right on Tinder, we have another masterpiece involving the matchmaking network and coding.
A Canadian engineer Justin Long (who probably got reeeealy tired of swiping right) created a plugin for the matchmaking service that knows what kind of people you like and can, for you, swipe right and even send love letters.
“While my friends were getting sucked into "swiping" all day on their phones with Tinder, I eventually got fed up and designed a piece of software that automates everything on Tinder”, he said (although we don’t really believe he did it for his friends).
Anyway, the Tinderbox plugin experiment is built using the service’s API and employs Eigenfaces, the facial recognition algorithm.
The plugin first recognises the user’s romantic preferences, and then goes hunting.
"Tinderbox is a full Tinder solution that learns who you're attracted to using machine learning and also has a built-in bot that can start conversations," said Long.
The workflow is simple: The built-in bot first builds facial models using your likes and dislikes, and then he examines profile images and crops faces. Those faces are loaded into an "average" face representing choices. Eigenfaces are computed from average faces, after which the bot makes future selections based on Eigenface comparison.
“Comparisons are essentially k-nearest neighbour selection”, says Long.
Long said the app was "amazingly effective" in his tests, finding an appropriate match 70 per cent (21 matches) of the time, while setting the bot to chat on his behalf led him to land ten dates, The Register writes.