matt-sharp/pixiedust-spark-wml

Analyse Customer Data and Recommend Products with PixieDust

An example of a same-day grocery delivery service is used to show you how to:

1. Explore and analyze customer behavior information (such as demographics and shopping cart values)
2. Build a machine learning recommendation engine to encourage additional purchases based on past buying behavior

IBM Cloud and open source tools and services, including Jupyter Notebooks, Apache Spark and Watson Machine Learning are used.

Before you start

• Sign up for an IBM Cloud account

• When you are signed up click Create Resource at the top of the Resources page. You can find the resources under the hamburger menu at the top left:

• Search for Watson Studio and click on the tile:

• Select the Lite plan and click Create.
• Go back to the Resources list and click on your Watson Studio service that you find under Services and then click Get Started.

IBM Watson Studio

1. Create a new Project

• You should now be in Watson Studio.
• Create a new project by clicking on Get Started and New Project, or Create Project
• Give your Project a name.
• Select an Object Storage from the drop-down menu or create a new one for free. This is used to store the notebooks and data. Do not forget to click refresh when returning to the Project page.
• Click Create.

Parts

• Part 1: Static data analysis using Python, Apache Spark and PixieDust
• Part 2: Build a product recommendation engine

Credentials

• Analyze historical shopping data with Spark and PixieDust in a Jupyter notebook
• Build a machine learning recommendation engine to encourage additional purchases based on past buying behavior