Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Implement deep learning for a music recommender system #105

Open
the-ethan-hunt opened this issue Jan 26, 2018 · 14 comments
Open

Implement deep learning for a music recommender system #105

the-ethan-hunt opened this issue Jan 26, 2018 · 14 comments

Comments

@the-ethan-hunt
Copy link
Owner

Deep learning can be used for recommender systems for movies, books, and music. See the presentation for more details.

@uds5501
Copy link

uds5501 commented Jan 28, 2018

This is do-able. I want to know in what format can I have the data and what all it will include. if it's going to be purely user's command based, then one can use NLTK repository.

I basically want to know what all data I can have to figure out which ML algo is the best in this case

@the-ethan-hunt
Copy link
Owner Author

@uds5501 , you need to be more specific regarding your application. We are going to implement deep learning for three stuff:

  • Movies
  • Music
  • Books
    Each of the above topic might have a different algorithm or a different neural network. Similarly the dataset will be different.

@the-ethan-hunt the-ethan-hunt changed the title Implement deep learning Implement deep learning for a music recommender system Jun 22, 2018
@the-ethan-hunt
Copy link
Owner Author

Update: We will for now begin with a music recommender system.

@arunikayadav42
Copy link

@the-ethan-hunt can I work on this issue please :)

@uds5501
Copy link

uds5501 commented Jun 25, 2018

@the-ethan-hunt May I know what all data attributes we will be provided? And as for recommendation engines, are you looking for a model collaborative filter with random forest or something else?

@the-ethan-hunt
Copy link
Owner Author

@arunikayadav42 you are already working on #109 . Please complete the issue and then proceed on this one 😄
@uds5501 a collaborative filter model is a great idea indeed. Data attributes will be provided akin to the Netflix dataset; but instead for music.
Hope you get the idea 😄

@uds5501
Copy link

uds5501 commented Jun 26, 2018

@the-ethan-hunt Seen the data, I would get back to you soon as soon as I think I've created a plan for the same!

@uds5501
Copy link

uds5501 commented Jun 27, 2018

@the-ethan-hunt , have a look at this excellent Kernel . It uses User-User Rating Collaborative Filter with Singular Value Decomposition. We just need a way to save the model weights (It isn't as easy as it usually is in Keras), Any suggestions on that?

@the-ethan-hunt
Copy link
Owner Author

Assigned to @uds5501 .
P.S. You have not included the link to the kernel.

@Arunkumar0258
Copy link

Hello, I can work on making a content-based filtered movie recommender system for BENJI. I can make start by making a simple NLP based system and then transitioning into a Deep Learning approach if needed.

@the-ethan-hunt
Copy link
Owner Author

Assigned to @Arunkumar0258

@shruthi0898
Copy link

Hi, I would like to work on a books recommender. Can I please get assigned on this and get details of the dataset?

@the-ethan-hunt
Copy link
Owner Author

Book recommender system assigned to @shruthi0898

@amrutha3
Copy link

Hey there @the-ethan-hunt! I would like to work on the music recommender system provided the dataset is given. I have some knowledge on the implementation of a recommender system and the math behind it . I'll try to implement it in Python.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

6 participants