raulcastr / song-recommender

Song Recommender

Raul Castrillo Martinez

Content

• Project Description
• Workflow
• Links

Project Description

Python app that recommends songs from a 100k song database ( gathered from Spotify API & Webscrapping) based on the audio features of the user's favourite songs or artists using an unsupervised Machine Learning algorithm (K-Means).

Workflow

• Create a database from https://www.billboard.com/charts/hot-100 using webscrapping for the top 100 popular songs.
• Create a database using spotify api with more than 100k songs.
• Deploy an unsupervised Machine learning algorithm (K-means) for the spotify database songs with its audio features.
• Coding all the functions to make the app able to interact with the user and recommend songs based on his/her favourite songs.
• Error handling and final testing.

Links

Repository