From the course: Machine Learning and AI Foundations: Recommendations

What is a recommendation system?

From the course: Machine Learning and AI Foundations: Recommendations

Start my 1-month free trial

What is a recommendation system?

“

- [Narrator] A recommendation system is a computer program that helps a user discover products and content by predicting the user's rating of each item and showing them the items that they would rate highly. Recommendation systems are everywhere. If you've ever looked for books on Amazon or browsed through posts on Facebook, you've used the recommendation system without even knowing it. With online shopping, consumers have nearly infinite choices. No one has enough time to try every product for sale. Recommendation systems play an important role in helping users find products and content they care about without having to spend all their time digging through things they won't like. Behind the scenes, these systems are powered by a recommender function. A recommender function takes in information about the user and predicts the rating the user would give the product. If you can predict the user's rating for a product before the user even sees the product, that's very powerful. That means you can show the user only the things they would like the best and not waste their time with products they won't care about. Imagine you're browsing for an ebook to buy on your ereader. The online book store knows about your past book purchases and the ratings that you gave them. Based on that historical information, it tries to predict how you will rate every product in its library. Using these predicted ratings, the book store will show you the books that it thinks you'll enjoy the most. These are also the books that you are most likely to purchase. Recommendation systems enhance the user experience while providing more exposure to a larger part of the company's inventory.

Contents