Search and Relevance
- Hive UI Framework
- Legacy Search UI Framework
- Using a Query Pipeline
- Altering Search Results Before They Are Displayed in a Search Interface
- Displaying External Content in a Search Interface
- Using Analytics
- Getting Started with Coveo Machine Learning
- Using Sitecore Personalization with Coveo
- Query Performance Leading Practices
- Understanding How to Leverage Virtual Users With Coveo
- Using LINQ to Customize Queries
Getting Started With Coveo Machine Learning in Sitecore
One of the many advantages of using a Pro or Enterprise Cloud edition of Coveo for Sitecore is the possibility to use Coveo Machine Learning (Coveo ML). Coveo ML is a powerful analytics-based machine learning service that continuously analyzes search behavior patterns by leveraging Coveo Usage Analytics (Coveo UA) to determine the most relevant items related to any given query. Coveo ML can also provide proactive query recommendations to your users, which you can place either on a search page or anywhere on your website where recommendations are useful.
Enabling Coveo Machine Learning in your Coveo Organization
By default, when you install Coveo for Sitecore, query suggestions are already active in your Coveo organization. It will start learning right away from the usage analytics data provided by the Coveo for Sitecore solution. You can also create a relevance tuning model to improve the default relevance of your solution, and a recommendation model to offer relevant recommendations to users according to their search and visit profile.
For the models to learn properly, you need to provide analytics data. If you don’t have a high traffic on your search pages, the models will keep building until there’s enough data.
Using Coveo Machine Learning in Coveo for Sitecore Hive
Once the Machine Learning models have been created in the Coveo organization, the output can be leveraged in the Coveo for Sitecore Hive framework.
Automatic Relevance Tuning (ART)
ART is applied in a Coveo-powered search interface to reorder results based on their popularity (see Using Automatic Relevance Tuning in Coveo for Sitecore).
Query Suggestions (QS)
QS provides a type-ahead experience to users typing in a Coveo Search Box. The suggested content is based on successful queries performed by past users. See Coveo Machine Learning Based Suggestions for a step-by-step guide on how to enable them in Coveo for Sitecore Hive.
Event Recommendations (ER)
ER can be used to recommend articles, products, and other related content to your users. The recommendations are provided by the ML model based on past users’ behavior. See Getting Recommendations for a step-by-step guide on how to enable them in Coveo for Sitecore Hive.