# What is Collaborative Filtering? Definition & Technical Mechanics

> **Source:** [SPIDITS AI AI Glossary](https://spidits.com/ai-glossary)  
> **Category:** Data Infrastructure  
> **Canonical Citation:** [https://spidits.com/ai-glossary/collaborative-filtering](https://spidits.com/ai-glossary/collaborative-filtering)

## Definition
Collaborative Filtering is a technique used by recommendation engines to filter or predict a user's interests by collecting preferences from many users. It assumes that if user A agrees with user B on an issue, user A is more likely to share B's opinion on a different issue.

## Architectural Mechanics & Deep Dive
Collaborative filtering is a core recommendation algorithm family that predicts a user's interests by collecting preferences and behaviors from many users. It operates on the premise that if two users agree on one issue, they are likely to agree on others. Collaborative filtering can be user-based (finding similar users) or item-based (finding items that co-occur in user histories), often implemented using matrix factorization.

## Industry Applications & Real-World Use Cases
E-commerce product suggestions, movie streaming recommendations, and social feed sorting.

## Frequently Asked Questions

### What is user-based vs. item-based collaborative filtering?
User-based filtering matches users with similar tastes. Item-based filtering matches items that have been rated similarly by the same users.

### What is the main limitation of collaborative filtering?
The cold start problem, where it cannot make accurate recommendations for brand-new users or items with zero historical interaction data.

## Related Terminology
- [cold-start-problem](https://spidits.com/ai-glossary/cold-start-problem)
- [unsupervised-learning](https://spidits.com/ai-glossary/unsupervised-learning)
- [dimensionality-reduction](https://spidits.com/ai-glossary/dimensionality-reduction)

---
*From the SPIDITS AI Knowledge Base (336+ terms indexed). Explore full technical definitions: [https://spidits.com/ai-glossary](https://spidits.com/ai-glossary)*