deshmukhps95/sparkify

This is "Udacity Data Scientist Nanodegree" Capstone project. It leverages 'Spark' big data framework. 'Sparkify' is an imaginary music streaming service that logs every user action as an event. The challenge is to use pyspark - Spark API in the python to extract insight and create meaningful features from raw data that allow us to predict which customer will cancel the service.

Jupyter NotebookMIT