Original article was published by Samkit Shah on Deep Learning on Medium
How Tinder is using the AWS Machine Learning service to find better matches through AWS Rekognition service
Tinder, One of the most popular dating application is using AWS Machine Learning Image recognition technology to power its matching for premium users.
Tinder is using AWS Deep Learning — Image Rekognition service to identify user’s key characteristics by mining 10 billion+ photos they upload daily. Tinder stores 40Tbs of data a day into its analytics and ML system to power matches, which are supported by AWS. Tinder knows that images are the primary driver for who you match. By analysis, the more picture you have the better the chances of getting matched. Just like the more data you have for your machine learning model the better the accuracy you’ll attain.
Tinder uses a Rekognition service which automatically tags billions of pictures that are uploaded by the users with a personality marker-like, a person with a guitar as a musician, a person climbing a mountain as adventurous, etc. In short, Rekognition provides Tinder with a way to “access what is inside these photos in a scalable way, that’s accurate and meets Tinder privacy and security needs.