This Blog is about Udacity Nanododegree Capstone project . I analyze user churn beahviour and create machine learning pipeline to predict churn.
The real relevance of the project is scaling Machine Learning by using Big Data technologie like Spark.
Sparkify is a digital music service (streaming) like Spotify and Deezer. Many users make subscribtion , listen their favourite songs (free or paid) with advertisement. Users can add friends, songs to their favourite list. Users can also upgrade or downgrade or even cancel subscriptions in any time.
For all streaming service, churn problem is the most relevant, so that it’s important…
Data scientist I am a problem solver with new technology