Skip to content

Latest commit

 

History

History
28 lines (15 loc) · 1.55 KB

File metadata and controls

28 lines (15 loc) · 1.55 KB

Build a custom recommendation engine with Amazon Personalize

This workshop will help you get started with the Amazon Personalize basic features in a short time. Please follow the steps in the Immersion Day ReadMe to set-up your account and evironment, copy relevant data and pre-train models if required.

Welcome to the UnicornPost

Congratulations! You have just been hired by UnicornPost, a publisher of news stories that captivate people the world over! Currently, most of your site does not provide a personalized experience, the stories are presented in a static order for all users in a manner controlled by the editorial team. In order to increase page views you are looking to incorporate a recommendation system which assists users in finding the stories they feel are the most intresting and informative.

You’ve been asked by the founders to:

  • Provide personalized story recommendation to vistors of your site
  • Filter news articles by genre for the relevant parts of the site
  • Emphasize newly written articles for certain sections of the site

In this workshop, we will add personalized recommendations to the UnicornPost SVOD website. The workshop is split into the following sections:

Get to know Amazon Personalize

Data Preperation and Import

Ready... Set... Train!

Let's take a look at and customize our recommendations

Clean Up Resources

Go through the sections one-by-one!