Introducing the Retailrocket dataset

In this chapter, we shall showcase a recommendation system algorithm using the Retailrocket dataset. 

Note

The Retailrocket dataset is available from the Kaggle website, at https://www.kaggle.com/retailrocket/ecommerce-dataset.

We download the dataset using the following command:

kaggle datasets download -d retailrocket/ecommerce-dataset

The downloaded files are moved into the ~/datasets/kaggle-retailrocket folder. You can keep it in whichever folder you feel comfortable with.

The Retailrocket dataset comes in three files:

• events.csv: This file contains the visitor-item interaction data
• item_properties.сsv: This file contains item properties
• category_tree.csv: This file contains the category tree

The data contains the values collected from an e-commerce website but has been anonymized to ensure the privacy of the users. The interaction data represents interactions over a period of 4.5 months.

A visitor can engage in three categories of events: view,addtocart, or...