- Built a recommender system based on content based filtering.
- Preprocessed Dataset of Tmdb with 5000 movies and created Tags for all movies.
- With help of scikit-learn Converted all tags to vector and calculated cosine distance of all vectors with respect to each other to find most familiar movies.
- Created a database on PostgreSQL to store all cosine distance and to fetch data for a movie whenever a recommendation has to be made.
- Used Render to host a web application for deploying Movie recommender system.
- Application fetches five most similar movies with respect to one entered and prints their names with respective posters.Posters are fetched with the help of Tmdb API.
click https://movie-recommender-0yro.onrender.com to try application.
Type movie name in the search box.
Then Click on recommend button.
Five most similiar movies will be shown to you.
