Learners

Matchbox recommender

This section describes the Infer.NET Matchbox recommender and is split into four parts:

• Introduction - gives a gentle introduction to recommender systems in general and briefly explains the core principles of the Matchbox model.
• Learner API - starts with a simple usage example of the Matchbox reocmmender and then goes in more detail on how to create the learner from a data mapping, how to save it and load it, how to change its settings, how to train it, how to make predictions, and how to evaluate those predictions.
• Model - explains the assumptions made in the Matchbox model and gives a full factor graph of the recommender.
• Command-line runners - describes how to use the Matchbox recommender from the command line.