This article describes how a Tealium Predict ML data model works and the components of a model.

In this article:

Table of Contents Placeholder

How it Works

Tealium Predict works with the Tealium AudienceStream CDP product to evaluate visit and visitor-level attributes. These attributes are used to create data models used to predict specific user actions.

Here's how it works:

  • Examine attribute readiness
  • Select a target attribute
  • Name your output attribute
  • Train the model and review
  • Select attributes to exlcude (optional)
  • Retrain and refine

Determining Attribute Readiness

The readiness of an attribute displays next to each attribute in the drop-down list of available attributes. This feature helps you determine, in advance, whether an attribute candidate is a Ready or Not Ready to use as a target output attribute, as shown in the following example.

Predict 1.1 Ready_Not Ready for Training.png

Target, Output, and Exclusion Attributes

When defining a model, each attribute in your AudienceStream profile is reviewed to automatically determine the top attributes that have a predictive relationship for the action you want to predict.

  • Target Attribute
    The target attribute is AudienceStream attribute selected to define your model and represent the user action being predicted. When building a model, you select a target attribute from the visit or visitor-level booleans attributes that display in the drop-down list.
    Target attribute are used to signal that an action has been performed. For example, a boolean visit attribute named "Has Purchased" signals that a purchase event has occurred during a visit.
  • Output Attribute
    The output attribute is an attribute created as a result of model training.
  • Exclusion Attributes
    You can select to exclude attributes that are not relevant for the results you are seeking. These are referred to as exclusion attributes. When deciding which attribute types to exclude, Tealium recommends that you first train the model for initial insights with no attributes excluded.
    Training without including exclusion attributes provides insight into which attributes the model finds the most relevant and can lead you to consider introducing new AudienceStream attributes to help future model trainings.
    After the initial training, you can exclude attributes with values that occur outside of the training period. After excluding these types of attributes, your training F1 score results may be lower when you retrain; however, your model produces better results when deployed.
    • Attributes based on dates of visit or dates of purchase. These attributes do not repeat their values outside of the training period.
    • Attributes based on unique user information, such as a User ID or Analytics ID. These attributes do not apply to other users outside of the training period.

See Advanced Configuration Options for the specific steps used to exclude one or more attributes from your model.

Get Started

Go to Adding a Model to get started.