Inspiration
As college students with exams and projects, we're too busy to go on Tinder. So we decided to automate the task in order to save our time for more productive activities.
What it does
RU-Single trains an AI to create a model to learn what kind of faces you're attracted to and then uses a bot to automatically login and swipe left or right on a profile depending on whether their face matches closely to the model.
How we built it
Model We first scraped 50 images from Tinder and classified them. This data set was used to train the Random Forest Model. We then used this model to predict classification. Bot Using Selenium, we created a bot to automatically login to Tinder, using Facebook, and we programmed the bot to either swipe-left or swipe-right depending on the result of the Random Forest Model's prediction.
Challenges we ran into
One challenge we ran into was eliminating images of unclear faces or pictures with multiple faces. Another challenge we ran into was getting the bot to recognize the swipe-left feature of Tinder as the webpage wouldn't load fast enough to keep up with the bot.
Accomplishments that we're proud of
The main accomplishment we're proud of is that the bot and model works. The model is able to accurately pick photos based off preferences and the bot had no trouble reading the image and swiping left or right depending on the Random Forest Model.
What we learned
We learned how to build classification models and bots with Python.
What's next for RU Single?
We'd like to improve our facial recognition feature to be able to recognize unclear faces and filter profiles by keywords in descriptions.
Log in or sign up for Devpost to join the conversation.