Machine Learning - Section

MachineLearningSection

The Machine Learning section of the Coveo Cloud administration console navigation menu allows you to manage and evaluate your Coveo Machine Learning (Coveo ML) (ML) models.

From this section, you can implement Machine Learning models and test their relevance impact on your search solution:

Coveo ML models are managed from the Models page only in Coveo Cloud organizations created after April 23, 2019 or that have been through the Coveo ML migration process. Other organizations manage models directly from the Query Pipelines page of the Search section (see Adding and Editing Coveo Machine Learning Automatic Relevance Tuning Models in a Query Pipeline, Adding and Editing Coveo Machine Learning Query Suggestions Models in a Query Pipeline, or Adding and Editing Coveo Machine Learning Event Recommendations Models in a Query Pipeline).

“Models” Page

On the Models page, you can manage Automatic Relevance Tuning (ART), Query Suggestions (QS), Event Recommendations (ER), and Dynamic Navigation Experience (DNE) ML models associated to a specific Coveo Cloud organization. This page allows you, among other things, to:

  • Create, edit, and delete Coveo ML models
  • Manage the learning interval for which the Coveo ML models gather data
  • Review Coveo ML model information

For further information on Coveo ML models, see Adding and Managing Coveo Machine Learning Models.

DNE models are only available for organizations created after May 20, 2019.

“Model Testing” Page

On the Model Testing page, you can compare two Coveo Machine Learning models of the same type to measure their effectiveness. This tool also allows you to compare the search results ranking provided by default with those provided by an active ART model.

For further information on model testing, see Testing Coveo Machine Learning models.