Implemented two types of recommender systems:
- Demographic Filtering - User gets recommendations based on the movie popularity.
- Content Based Filtering - User gets recommendations based on the movie's metadata like director, actors, genre, keywords etc.
- Python libraries like NumPy, Pandas and Matplotlib are used to manipulate and visualize data.
- TF-IDF, linear kernel and cosine similarities from scikit-learn library are used to generate similarity score matrix to recommend the most relevant movies.
- Detailed step by step explanation can be found in the attached code (Movie_Recommender.ipynb)
- TMDb dataset of about 5000 movies (attached)