ManySpace
Türkçe Sign in Sign up

AI & MLRecommendedAIML-REC-003

Matrix Factorization for Recommendation

Users and items as vectors in a shared latent space; ratings as dot products.

Difficulty
3 / 5
Importance
4 / 5
Study time
1–2 h
Kind of work
Mixed

Why it matters

The learner can vary the latent dimension of a factorised rating matrix and explain the underfitting/overfitting trade-off on held-out ratings.

The route to this topic

Route AIML-REC-003
What Is a Model? Tasks: Regression, Classification, and Beyond Features and Targets The Recommendation Problem Linear Regression I: Prediction and Residuals Embeddings Loss and Fitting: Squared Error Similarity in Vector Space Linear Regression II: Least Squares Fitting Collaborative Filtering Principal Component Analysis Matrix Factorization and the SVD Matrix Factorization for Recommendation