Skip to content

romy651/Sparkify

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Sparkify Project: Predicting User Churn

Project Overview

Sparkify, a fictional music streaming startup, aims to predict user churn using Apache Spark. In this context, "churn" refers to customers leaving the service over a given period. Identifying potential churners is crucial, as acquiring new customers can cost five to six times more than retaining existing ones.

Go to blog post!

User Types and Churn Definition

  • Free plan users: Stream music with ads
  • Paid plan users: Ad-free music streaming
  • User actions: Thumb up/down songs, create playlists, add friends
  • Churn types:
    1. Downgrade: Paid to free plan
    2. Cancel: Stop using the service entirely

Project Goal

Develop a machine learning model to accurately identify potential churners of both types (downgraded and cancelled users) based on their interaction data.

Key Points

  • Data: 12GB log file with 18 fields per user interaction
  • Tools: Apache Spark, PySpark MLlib
  • Process: Data loading, exploration, feature creation, model building, churn prediction
  • Models Tested: Logistic Regression, Support Vector Machine and Gradient Boosted Trees
  • Best Model: Tuned Logistic Regression

Project Significance

  • Addresses critical business challenge of customer retention
  • Demonstrates handling of large datasets with Spark
  • Applies machine learning techniques to real-world problem

Results

The winning model effectively identifies users at risk of churning, enabling targeted retention strategies to potentially save millions in revenue.

About

Sparkify Project: Predicting User Churn

Resources

Stars

0 stars

Watchers

1 watching

Forks

Releases

No releases published

Packages

 
 
 

Contributors