https://github.com/jaimescose/music_recommender

https://github.com/jaimescose/music_recommender

Last synced: 28 days ago
JSON representation

• Host: GitHub
• URL: https://github.com/jaimescose/music_recommender
• Owner: jaimescose
• Created: 2024-10-28T02:44:28.000Z (about 2 months ago)
• Default Branch: main
• Last Pushed: 2024-10-28T02:49:38.000Z (about 2 months ago)
• Last Synced: 2024-10-28T06:24:54.607Z (about 2 months ago)
• Language: Python
• Size: 852 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Artist Recommender System

## Dataset

Original dataset can be foundd [here](https://grouplens.org/datasets/hetrec-2011/).

However, for convenience, I have uploaded just the files that I used in this project: 

- artists.dat: matches artist id to name and other information (not relavant for this project)

- user_artists.dat: matches user id to artist id and weight (representing the preference of the user)

## Model

This basic music recommender system uses the library [Implicit](https://github.com/benfred/implicit),

more specifically, the Alternating Least Squares model (Collaborative Filtering for Implicit Feedback Datasets).

This model takes as input an Sparse Matrix (CSR) of user-artists interactions.

## Usage

Uses [uv](https://docs.astral.sh/uv/) to describe project dependencies.ç

## Future Work

Be able to access Spotify's data to get the user-artists interactions.