Yahoo Music Recommendation System

In this project, we built a Recommender System based on the hierarchical music data set of Yahoo.

Dataset Description and the competition

This is the final challenge of the EE627 course. Nowadays, technology companies often use recommender algorithm to recommend products, music, movie, etc. Recommender algorithm is really useful and helps all those companies make a huge profit. Now, after learning the factorial matrix and other useful algorithms. It's our turn to solve the problem which Spotify, Amazon, and Yahoo Music faced every day - recommend music to their customers.

Overview

• trainIterm2.txt - the training set
• testIterm2.txt - the test set
• sample_ submission.csv - a sample submission file in the correct format
• trackData2.txt -- Track information formatted as: <'TrackId'>|<'AlbumId'>|<'ArtistId'>|<'Optional GenreId_1'>|...|<'Optional GenreId_k'>
• albumData2.txt -- Album information formatted as: <'AlbumId'>|<'ArtistId'>|<'Optional GenreId_1'>|...|<'Optional GenreId_k'>
• artistData2.txt -- Artist listing formatted as: <'ArtistId'>
• genreData2.txt -- Genre listing formatted as: <'GenreId'>

Link to the Competition

https://www.kaggle.com/c/ee627A-Spring2019/overview