Monthly Archives: April 2025

One day I was drinking coffee with my friend and he told me a story of how he got in trouble with his girlfriend because she found his old profile on a popular dating site match. Allegedly someone from their common friends bumped into it and sent her a friendly notification. He was unable to prove her that it’s just an old thing’ that he forgot to delete and their story ended there they broke up pretty fast after this incident. What are the odds of bumping into a profile of someone you know on a dating website and how easily such privacy could be violated if someone had a direct intention?

I remembered that there were some open-source face detection and recognition libraries available and thought that it’s probably possible to write a tool that would crawl photos on dating sites and try to recognize a particular person on them. Then I ran into face a platform that provides a RESTful API for detecting, tagging and recognizing faces on the pictures. I recalled that story again, told to my friends, we all laughed, agreed that such a tool would be creepy, but I did put it in my list of ideas for hacking.

So, guess what, let’s go creepy and run a small experiment to see how easy that would be. To do that we’ll write a tool that will take tagged photos of a Facebook user and try to find his/her profile on match.

• How send authorized search requests to match
• How to get URLs of profile images
• How to make parsing fast by running requests asynchronously
• How to use face API

Sending authorized requests

To get profile pictures we need a search output. If you go to match you’ll figure that it doesn’t allow you to browse search results without registration. So the very first step would be to create an account on match. After that go to the search form, choose the parameters, like sex, age and a zipcode and click Search now.

Continue reading