Learn Recotem
Task-focused guides for building real recommendation systems with Recotem. Each guide is self-contained and grounded in a runnable recipe — pick the one that matches the data you already have, or the decision you are trying to make.
Basics
New to recommender systems? Start with the fundamentals.
- What Is a Recommendation Engine? — how recommenders work, the main algorithm families, and how to get started.
- Collaborative Filtering Explained — user-based, item-based, matrix factorization, and implicit vs explicit feedback.
Use cases
Build a working recommender from the data you already have.
- Product Recommendations from GA4 + BigQuery — turn Google Analytics 4 events into a live recommendation API.
- Recommendations from Purchase Logs — "customers who bought this" from an e-commerce order history.
- Recommendations from a SQL Database — train directly from PostgreSQL, MySQL, or SQLite.
- Add a Self-Hosted Recommendation API — serve recommendations over HTTP from your own infrastructure.
- E-commerce Product Recommendations — placements, data, and measurement for an online store.
Compare
Where Recotem fits relative to the alternatives.
- Build vs Buy for Recommendation Engines — Python libraries vs SaaS vs open-source self-hosted.
- AWS Personalize Alternative — a self-hosted, open-source option.
- Recotem vs LightFM vs implicit — framework vs library for Python recommenders.
- Open-Source Recommendation Systems Compared — Gorse, RecBole, Merlin, and Recotem.
Next steps
- New to Recotem? Start with the Guide.
- Need field-level details? See the Recipe Reference.