Concepts

Almeta ML uses customer action data to predict future behavior. This data is collected via Events, which are used to train models and make predictions.

 

Events

Events are small pieces of data that represent actions that customers take on your website. For example, a purchase, a page view, or a form submission. Events can be imported using web tags, integrations, API, or data import.

TIP

Learn more about Events.

Models

Models are machine learning algorithms that are trained on customer action data to predict future behavior, like likelihood to purchase, churn, or finish a course.

When you create a model, you can select conditions and triggers to run the calculation (inference). For example, you can create a model that predicts the likelihood of a customer making a purchase after viewing a number of website pages.

Destinations

Destinations are places where the results of ML model calculations are sent. For example, you can send predictions to Facebook Ads, Google Ads, or Shopify. Almeta ML runs machine learning model calculations and sends results to selected destinations.

Catalog Data

Catalog data is a list of products or services that you want to use in your predictions. For example, you can use a list of products to predict the likelihood of a customer purchasing a specific product.