/linkedin-recommendation-systems-a-practical-introduction-2703578

This is a repository for the LinkedIn Learning course Recommendation Systems: A Practical Introduction

Primary LanguageJupyter NotebookOtherNOASSERTION

Nightly builds

Recommendation Systems: A Practical Introduction

lil-thumbnail-url

This is the repository for the LinkedIn Learning course Recommendation Systems: A Practical Introduction. The full course is available for free from LinkedIn Learning.

Recommendation systems are among the most profitable Artificial Intelligence solutions you can deploy, for the simple fact that they can understand what people are interested in. Anytime you buy or browse online, there are probably recommendation systems at work presenting you with options at each step.

In this free course, Miguel Fierro teaches a practical introduction for building, deploying, and testing recommendation systems. He offers practical, real-world examples to show how you can make a direct impact in your company with recommendation systems.

Whether you’re a Data Scientist, Machine Learning Engineer, Data Engineer, Software Engineer, or Data Analyst, join Miguel in this course to get started building your first recommendation system.

Content

The course has the following sections:

  1. Introduction
    • Why recommendation systems?
  2. Data Prep
    • Data in recommendation systems.
    • Data splitting.
    • The cold-start problem.
  3. Modeling
    • Recommendation systems algorithms.
    • Collaborative filtering.
    • Content-based filtering.
    • Building your first collaborative filtering solution.
    • Building your first content-based filtering solution.
    • Evaluation of recommendation systems.
  4. Deployment
    • Recommendation system architectures.
    • Evaluation in production.
  5. MLOps
    • Tests in recommendation systems.
    • The machine learning lifecycle.

Code

Installation via GitHub Codespaces

  1. Go to Code.
  2. Click on Create codespace on main.
  3. After the codespace is ready, you can run all the notebooks.

Installation in local

conda create -n reco Python=3.7
conda activate reco
pip install numpy "Cython<4" "scipy<1.11.0"
pip install -r requirements.txt

Test

To test the code, run the following command:

pytest tests