Helping Train Coveo Machine Learning With Thesaurus Rules

With the arrival of Coveo™ Machine Learning (Coveo ML), habitual ways of improving search results relevance such as the creation of thesaurus rules become less necessary. Indeed, Coveo ML can, among other things, learn synonyms and words behind acronyms over time simply by analyzing the usage analytics data of a specific search interface. However, you can help accelerate the Coveo ML training process by pointing to Coveo ML synonyms as well as words behind acronyms.

The help you provide lasts as long as the data is used to train the Coveo ML model in the worst case scenario (see Data Period). Best case scenario, your search interface users repeat the actions you take to train the model.

You point to Coveo ML that BO means business optimization and your ART model is based on three months of data. In 90 days, you may have to help Coveo ML understand the same thing if no users searching for BO and business optimization clicked the same results.

To help train Coveo Machine Learning with thesaurus rules

  1. Ensure Coveo Machine Learning features are configured and enabled on your search interface (see Managing Coveo Machine Learning Automatic Relevance Tuning Models in a Query Pipeline and Managing Coveo Machine Learning Query Suggestions Models in a Query Pipeline).

  2. In your search interface, in your search box, enter the acronym or a term, and then click an appropriate result.

  3. Enter the words behind the acronym or a synonym of the term entered in the previous step, and then click the same result.

  4. Repeat the procedure for each acronym or synonym.

  5. Once the model is trained, in the Content Browser, ensure the model returns the expected results. Repeat the procedure if needed.

    It may take up to one hour before your model recommends results or suggests queries based on your training.