For many users, the idea of sorting through hundreds of dating profiles to find potential matches seems daunting and time-consuming. This automated system recommends profiles of other users that a user will like! We're turning a cold-start problem into a user to item problem by calculating similarity vectors between new and existing users' data. We then apply additional filtering and also TF-IDF for the text-based inputs to recommend profiles based on the comprehensive method.
Clone this repo to your local machine
pip install -r requirements.txt
streamlit run app.py
