What is diversity in recommendation systems?

Updated May 16, 2026

Short answer

Diversity ensures recommended items are varied and not too similar.

Deep explanation

Diversity measures how different recommended items are from each other in terms of content, category, or embeddings. High accuracy alone can lead to repetitive recommendations. Diversity is often introduced using re-ranking, submodular optimization, or penalties for similarity.

Real-world example

Spotify mixing genres instead of only one type of music.

Common mistakes

  • Optimizing only accuracy and ignoring user experience.

Follow-up questions

  • What is trade-off between diversity and relevance?
  • How is diversity measured?

More Recommendation Systems interview questions

View all →