Skip to main content

Recommender systems and examples

Short answer

A recommender system filters information to suggest items by collecting user data, building profiles, predicting preferences via content-based or collaborative methods, and ranking top items. Examples: Amazon 'customers also bought', Netflix video suggestions, Spotify Discover Weekly.

What is a recommender system? How does it work? Discuss different examples of recommender systems.

A recommender system is an information filtering system that predicts the preference or rating a user would give to an item and suggests the most relevant items (products, movies, news, friends) to the user. It reduces information overload.

How it works

 User data + Item data
        |
   [Collect data]
        |
   [Build profile / model]
        |
   [Find similarity / predict]
        |
   [Rank and recommend top-N]
        |
   [Feedback -> update]
  1. Data collection: explicit ratings and implicit behaviour (clicks, purchases, watch time), item features and user profile.
  2. Learning: the system builds a user profile and item model.
  3. Prediction: it uses a technique such as content-based (item similarity to past likes), collaborative filtering (similar users), or hybrid.
  4. Recommendation: it ranks items and shows the top few.
  5. Feedback: the user's reactions improve future recommendations.

Examples

  • Amazon: "customers who bought this also bought" (item-based CF).
  • Netflix / YouTube: recommended videos using watch history and hybrid methods.
  • Spotify: playlists such as Discover Weekly.
  • Facebook / LinkedIn: "people you may know".
  • Google News / Daraz: personalised news and products.
  • Booking sites: hotels based on previous searches and location.

✓ Answer checked by Claude Opus 5.5 (high effort)

Related questions

All Information System chapter 7 questions →