1. Narrative Knowledge Base
  2. Integrations & Technical Guides

Rosetta Stone Attribute Mapping Management UI

How to use the Platform UI to see, edit and create Rosetta Stone Attribute mappings.

About Rosetta Stone Attribute Mappings

Mappings play a pivotal role in standardizing data from various sources. They translate and normalize the data, ensuring consistency and interpretability for all collaborators and easing the process of data interaction and utilization. Mappings are applied to datasets. Each mapping links a specific Rosetta Stone Attribute to a field in a dataset, allowing for clear data categorization. Each mapping also defines a transformation, ensuring that each data point is represented consistently with the definition of the attribute. To learn more about Rosetta Stone Attributes, read the knowledge base article How Rosetta Stone works.

Attributes Tab: Your Hub for Attribute Mappings

  • Location: Find the “Attributes” tab easily within the Dataset Details section of the Platform UI when viewing any dataset that you own.
  • Functionality: This tab is your primary interface for viewing and managing Rosetta Stone Attribute Mappings.
  • Features:
    • Searchability and Sortability: Quickly find the specific mapping you need.
    • Expandable Rows: Delve into the details of each attribute mapping.
    • View/Edit Mapping Expressions: Directly interact with and modify mapping expressions, which describe how a field is transformed into an attribute.

Viewing and Editing Mapping Expressions

Use the action menu to interact with existing and proposed Rosetta Stone attribute mappings. 

Creating New Attribute Mappings

Click "+Add Mapping" in order to begin creating a new attribute mapping. You will see sample data as well as the option to save, test your new mapping, or cancel.

Attribute mapping expressions must be valid NQL. Additionally, attribute mapping expressions should generally reference one or more fields in your underlying dataset schema.

It is a best practice to escape the name of your schema fields with double quotations marks (") to avoid conflicts with NQL reserved keywords. 

Read more about NQL here: Narrative Query Language Overview.

Automatic Dataset Mapping with Rosetta AI

Clicking "Ask Rosetta" will enable our automated mapping feature for the dataset. Our automated mapping functionality applies advanced machine learning techniques to evaluating your dataset based on the names and types of the fields in the schema of your dataset, as well as a sample of data elements themselves.  

Rosetta will create all new attribute mappings in the state "pending" .  In order to utilize a new attribute mapping, you must "accept" the mapping. (See states below). 

Attribute Mapping Statuses

In the Platform UI, you will encounter several fields when managing your Rosetta Stone attribute mappings. These fields - scope, source, and state - play crucial roles in defining the accessibility, origin, and status of each mapping. Here’s a breakdown to help you understand what each field represents:

Scope

The scope field determines the reach or applicability of a mapping.

  • private: This scope is limited to company-specific queries. It means the mapping is visible and applicable only within your company's domain. Users can only create private mappings. 
  • global: Mappings with this scope are accessible to any query targeting the dataset. Only Narrative administrators or the system itself can set a mapping to this scope.

Source

The source field indicates who or what created the mapping.

  • admin: This mapping was created or promoted from a private mapping by a Narrative administrator.
  • company: If you see this source, it means the mapping was manually created by a user like you on the platform.
  • system: This signifies that the mapping was auto-generated by the Narrative system.

State

The state field reflects the current status of a mapping.

  • active: An active state means the mapping is currently in use and can be applied in access rules, or when selecting fields directly from the dataset using Rosetta Stone (`_rosetta_stone`). 
  • pending: This is a preliminary state. Mappings in draft need your approval to become active and usable in access rules and when querying the dataset directly using Rosetta Stone. 
  • rejected : A rejected mapping is not to be applied or used. Additionally, Rosetta will not attempt to create a new mapping for this specific attribute to this dataset.
  • deleted: Archived mappings are inactive and are not returned in the API and will not be visible in the UI. However, Rosetta can attempt to apply a new mapping for the given attribute.