You are here:
Use Clustering on Structured Data
Use clustering models in Data 360 to automatically group data model object (DMO) records by shared characteristics. Run a predict job or batch data transform to assign new records to saved clusters, and use the results across Salesforce.
Required Editions
| Available in: All Editions supported by Data 360. See Data 360 edition availability. |
| User Permissions Needed | |
|---|---|
| Allow users to manage models in Einstein Studio (AI Models) | Enables you to create, update, and delete models in the Data 360 AI Models tab. |
| Permission sets | |
|---|---|
| Data Cloud Architect | Admin-level access to all AI Models features, including the ability to create, update, delete, and activate models. |
| Data Cloud User | Restricted access to use a model, including getting predictions and improvements derived from a model. |
Before you begin:
- Build and activate a clustering model in the AI Models tab in Data 360.
- Make sure that you use a DMO with structured data.
Tips:
- Use scores to rank or filter records by confidence. If a record has a low cluster score, it can be a poor fit for its assigned cluster.
- Cluster label names reflect what you set during model setup. To change labels, edit the clustering model and rerun the transform.
Map Records to Clusters with a Predict Job
Use a predict job to run inference and map new records to clusters in a DMO.
- From Data 360, go to the AI Models tab, and click the activated clustering model.
- On the Integrations tab, select Add Job. Predict Job Builder opens.
- Set the model inputs using a DMO, and map its fields to the model’s variables.
- Define the output DMO. Optionally, specify how many top contributors to include.
- Schedule a streaming or batch job.
- Save the job.
- Activate and run the predict job.
Run Clustering with a Data Transform
Use a batch data transform to apply cluster labels to new data in a DMO.
Important
Deleting the clustering model removes all downstream transform jobs. Before you delete the model, delete all associated transforms.
- From Data 360, go to the AI Models tab, and click the activated clustering model.
- On the Integrations tab, select Add Transform.
- In the transform setup, select the structured DMO that contains the records you want to assign to clusters.
- Map the DMO fields to the corresponding fields used during model training.
-
Configure the output DMO settings.
- DMO name: The name of the output DMO.
- Cluster ID: The name for the cluster ID field (the API name is set automatically and you can't change it).
- Cluster Label: A descriptive label that summarizes the records in the cluster.
- Configure up to 3 top contributors to include in the output.
- Activate and run the predict job.
- Save and run the transform.
¿Resolvió este artículo su problema?
¡Háganos saber cómo podemos mejorar!
