Skip to content

Latest commit

 

History

History
10 lines (8 loc) · 1.92 KB

README.md

File metadata and controls

10 lines (8 loc) · 1.92 KB

Recommender-in-Azure

Accompanying material to the course Building Recommendation Systems in Azure in the Microsoft Virtual Academy.

The course is built as follows:

  1. Machine Learning & Recommenders: This module gives a short introduction in machine learning and recommendation systems. It particularly highlights the two main approaches of recommendations - collaborative and content-based filtering.
  2. Targeted Marketing: Before digging into recommendation systems, we look into the most simple form of machine learning problems - binary classification. Classification in general could also be used as a basis for recommendations.
  3. Collaborative Filtering: Association Rules in R & AzureML: Here, we go through one common recommendation approach using RStudio. Then we integrate the R script into Azure Machine Learning
  4. Content-Based Filtering & Hybrid: Matchbox Recommender: We use the built-in machine learning algorithm in AzureML (short for Azure Machine Learning) - MatchBox recommender. In its basic form it is a rating-based/content-based recommendation approach. This can be extended to a hybrid recommender by integrating user as well as item features.
  5. Recommendations API (Azure ML Marketplace): What to do if I cannot be bothered to build a machine learning model but still want to give personalised recommendations? Don't worry - this is where machine learning APIs, such as the Recommendations API are provided.