Recommendation Systems Theory and Coding

This advanced path develops the theory and implementation skills needed to engineer recommendation systems in Python. It begins with baseline prediction models, similarity measures, and the foundations of user-based and item-based collaborative filtering.

You will progress to matrix factorization, Alternating Least Squares, and implicit-feedback techniques. You will also build content-based approaches using item features, user profiles, Factorization Machines, and Deep Structured Semantic Models.

The path concludes with methods for assessing recommendation quality beyond prediction accuracy. It is intended for experienced programmers who want to design, implement, and evaluate personalized recommendation models.

Syllabus

Implement baseline predictors and similarity measures in Python
Build user-based and item-based collaborative filtering models
Apply matrix factorization and Alternating Least Squares to recommendation tasks
Model implicit feedback for personalized recommendations
Develop content-based models using Factorization Machines and DSSM
Evaluate systems using coverage, novelty, diversity, and serendipity metrics