Recommender systems


Recommender systems

Training description

Mechanisms that suggest the next movie to watch or a product that we might want to add to the cart are nowadays widespread and well known to anyone who uses the internet. Recommender systems are what powers those mechanisms. This workshop covers the process of building such system from scratch. The most popular algorithms along with their pros and cons are covered.

Duration: 3 days

Requirements: knowledge of Python programming language at an intermediate level (especially in data processing field).

Training agenda

Part one: Introduction

  • What is a recommender system
  • Applications of recommender systems

Part two: Algorithms

  • Content-based recommendations
  • Association rules
  • Nearest Neighbor Collaborative Filtering
  • Matrix factorization
  • Factorization Machines and Field Aware Factorization Machines
  • Introduction to neural network in recommender systems

Part three: Evaluation of recommender systems

  • Mean Average Precision at K (MAP@K)
  • Mean Average Recall at K (MAR@K)
  • Coverage
  • Personalization
  • Intra-list Similarity

Part four: Extensions to recommender systems

  • Context awareness
  • Risk awareness

Part five: Challenges in recommender systems

  • The cold start problem
  • Scalability
  • Interpretability

Contact us about closed training

This website uses cookies to ensure you get the best experience on our website.
Ok, got it. More about cookies