How to Build a Recommendation Engine
Between 2011 and 2012, recommendations account for 35% of all sales on Amazon. Similarly, Facebook, LinkedIn, Nextdoor, Gilt, and Netflix all heavily employ recommendation engines in their software.
Henry explains two approaches based on collaborative filtering: user-based, using a simple library called GER (Good Enough Recommendations), and item-based. He then shows a live demo of Hipflix, a recently deployed VHS eCommerce store he built with his team in a week.
Project Members: Henry Bao-Viet Nguyen