Skip to content

Latest commit

 

History

7 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 

Repository files navigation

VASP-Recommender-System

A Tensor-flow implementation of a Recommender System which takes in a sparse user-item rating matrix to generate new entries using which items can be recommended to users.

It is inspired by the model described in the publication: Deep Variational Autoencoder with Shallow Parallel Path for Top-N Recommendation (VASP) It is recommended to read the above paper for better unerstanding of the implementation.

Note: A closed-form solution of EASE is implemented here instead of the Neural EASE model.

Contributors:

Uthpal P Siddhanth M Sathvik Rao

Implementation Description:

Input:

A sparse matrix containing the ratings of movies rated by different users obtained from MovieLens100k.

Output:

A Hadamard product of the matrix generated using the EASE Model and the matrix generated using a Varational Auto-Encoder. This matrix contains new user-movie rating entries which is used to provide new movie recommendations to the users.

About

A TensorFlow Implementation of a Recommender System which makes use of Variational Auto-Encoder(VAE) and EASE model.

Resources

Stars

1 star

Watchers

1 watching

Forks

Releases

Packages

Contributors

Languages