This is an implementation of two popular recommendation techniques (collaborative filtering and latent factor model) based on the Mining of Massive Datasets video series. In this implementation we work with predicting anime ratings using the CooperUnion Kaggle anime dataset. This project was a collaboration between Scott Freitas and Benjamin Clayton.
(1) Create three folders in the code directory named 'csv', 'matrices' and 'optimization'
(2) Download the anime dataset from Kaggle: https://www.kaggle.com/CooperUnion/anime-recommendations-database
(3) Add the 'anime.csv' and 'rating.csv' files to the 'csv' folder
(4) Run the 'RecommenderSystem.py' file and it will walk you through the process of running the program with an interactive dialogue.
This program was built to run with Python 2.7 in a Windows environment. In additions, we used the numpy, scipy and sklearn libraries.
