Summary

I have the responsibility of developing a model to identify users who are likely to downgrade from the premium to the free tier or stop using the service entirely. For the purpose of creating machine learning models with massive datasets, I'll utilize Spark MLlib and scikit-learn.

A digital music service called Sparkify is comparable to Spotify, Apple Music, and Youtube Music. Users have the option to subscribe, add friends or songs to their playlists, listen to their favorite music for free (with advertisements) or for a fee,...

Prerequisites

    pip install -r requirements.txt

Run the notebook Sparkify.ipynb

The findings of this project already published follow the link:

https://medium.com/@quanvu.hcmvb220204114/data-science-final-project-c10070fab9e3

Data

mini_sparkify_event_data.json