Case Study: Recommendation Engine for Movies

Thanks for signing up to receive this free case study extracted from MIT's online course on data science.

Download Case Study


Recommendation Engine For Movies: Using Data to Provide The Best User Experience

Instructor: Devavrat Shah
Course: Data Science and Big Data Analytics: Making Data-Driven Decisions 

Ever wonder how industries like Netflix, Spotify and Pandora filter products based on their unique user’s preference? Using data of course! But how do they get that data? By building what is known as a Recommendation Engine – a feature that filters items by predicting how users will rate them – the goal is to connect users to the right items so that they will continue to use products/services.

In this case study, we will focus on Netflix and how they utilize Recommendation Engines to provide the best possible shows and movies for unique users.

IMPORTANT: Don't get discouraged if some of the steps described seem too complicated! Remember, this is an extract of the online course that will provide you with all the background necessary to successfully complete this activity.


  1. Watch the video under "Watch & Learn" ( Recommendation Engine for Movies Video) — it's taken from the course and provides background knowledge you will need to complete the case study.
  2. Download the case study and follow the directions inside the case study.
  3. Eager to learn more? Stay tuned to receive more information about the course soon!

Download Case Study

In this case study, you will learn

  • How to build your very own Recommendation Engine.
  • Analyze and segment data
  • Data Splitting
  • Using tools such as RecommenderLab and GraphLab-Create to find popular items.
  • Filtering items together based on similarity in popularity and content.
  • Collaborative filtering – using data and feedback from past users to provide the best possible experience for current users.
  • Top-K recommendations – setting up an algorithm that provides top recommendations based on user’s current preference (provides items that have not been rated by a user).
  • Get a sneak peek at the content included in MIT's online professional course on data science.

Watch & Learn