The Letterboxd Recommender is a project aimed at providing personalized movie recommendations based on user preferences and ratings. It utilizes the Letterboxd API to fetch movie data and user ratings.
I will use this data to compare and implement different recommendation algorithms:
- Random Walks
- Transformer based
- Re-scrape user if last data too old. Can also always check that we have the user's latest movies and just scrap the latest
- Constants for data paths and such
- Testing the graph model embedding quality