Music Recommandation System based on Yahoo! music dataset.
-
Hierarchy.py
generate hierarchy structure for a list of track items. -
Rating.py
generate ratings for all the items in the hierarchy structure. -
matrix_factorization.py
apply matrix_factorization on datasets to make predictions. -
ADD.py
make predictions by simply adding album and artist rating. -
Ensemble.py
ensemble all the results obtained from previous steps and make new predictions. -
Report.pdf
introduction of methods applied in this project.
Here is a instruction about how to use the programs. All the methods are mentioned in the report.
Methods 1--6
-
Run
Hierarchy.py
to get the hierarchy structure. -
Run
Rating.py
to get ratings for every item of each user. -
Run
Add.py
to apply our methods by changing some parameters.
Method 7
Run matrix_factorization.py
to perform matrix factorization method by adjusting rank and numiter
Method 8
Run Ensemble.py
to perform ensemble method