Manage semantic types for fields - Precisely Data Integrity Suite

Data Integrity Suite
Product
Spatial_Analytics
Data_Integration
Data_Enrichment
Data_Governance
Precisely_Data_Integrity_Suite
geo_addressing_1
Data_Observability
Data_Quality
dis_core_foundation
Services
Spatial Analytics
Data Integration
Data Enrichment
Data Governance
Geo Addressing
Data Observability
Data Quality
Core Foundation
ft:title
Data Integrity Suite
ft:locale
en-US
PublicationType
pt_product_guide
copyrightfirst
2000
copyrightlast
2026

Semantic types classify the meaning and role of data fields. You can view built-in semantic types, assign them to fields, create custom semantic types from profile data, and update or remove assignments.

A semantic type is a category that classifies the meaning or role of a concept, entity, or data field within a specific context. Rather than focusing on structure or syntax, semantic types group items based on what they represent or signify. Semantic types are determined by their associated semantic type qualifiers, which are standardized codes or identifiers that classify data fields according to their semantic meaning and purpose.

Common semantic types: The following table lists commonly recognized semantic types with their qualifiers and examples:

Semantic type name Semantic type qualifier Description Example
Email Address EMAIL Identifies a string as an email address. Represents valid, commonly used real-world email values. user@example.com
Telephone Number (Generic) TELEPHONE Identifies a string as a phone number. Represents valid, commonly used real-world phone values. +1 415 555 9823
Country As A String (English) COUNTRY.TEXT_EN Identifies a string as a country code in English. Represents valid, commonly used real-world country codes. US
Social Security Number (US) SSN Identifies a string as a US Social Security Number. Represents valid, commonly used real-world SSN values. 123-45-6789
Postal Code (US) POSTAL_CODE.ZIP5_US Identifies a string as a US postal code. Represents valid, commonly used real-world postal code values. 94105
Street Address STREET_ADDRESS2_EN Identifies a string as a postal address in English. Represents valid, commonly used real-world address values. 221B Baker Street, London
First Name NAME.FIRST Identifies a string as a first name. Represents valid, commonly used real-world first name values. Michael
Last Name NAME.LAST Identifies a string as a last name. Represents valid, commonly used real-world last name values. Thompson
Merged Name (Last, First) NAME.LAST_FIRST Identifies a string as a name combining both first and last names. Represents valid, commonly used real-world name values. Michael Thompson
City or Town CITY Identifies a string as the name of a city. Represents valid, commonly used real-world city values. San Francisco
US State Code STATE_PROVINCE.STATE_US Identifies a string as the name of a US state. Represents valid, commonly used real-world state values. California

Semantic type states: A field can have one of the following semantic type states:

  • Not detected: No semantic type has been assigned to the field. This is the initial state before automatic detection runs or before you manually assign a type.
  • Detected type: A semantic type has been automatically detected and assigned to the field.
  • Manually assigned type with Overridden badge: A semantic type has been manually assigned, overriding any automatic detection. The system displays an 'Overridden' badge to indicate manual assignment.

Entry points: You can manage semantic types for a field from multiple locations:

  • Details tab: In the field details page, go to the Details tab and locate the Semantic Type field in the Summary section.
  • Profile tab: In the field details page, go to the Profile tab and locate the Semantic Type field in the Summary section.
  • Side panel: In the Fields tab of the Catalog, select a field to open the side panel and locate the Semantic Type field.

Assign a semantic type to a field

Manually assign a semantic type to a field when automatic semantic type detection has not run or when you want to assign a specific semantic type that differs from the detected value.

Manual assignment allows you to override the default semantic type and ensures the assigned type is used for rule detection and data quality analysis.

  1. Go to Catalog > Fields.
  2. Click the field name to open the field details page.
  3. Go to the Details or Profile tab.
  4. In the Summary section, locate the Semantic Type field and click the edit icon next to it.
  5. In the Update Semantic Type dialog, review the available options:
    • Detect Semantic Type: Runs automatic semantic type detection and updates the assignment. This option is useful if you want to re-run the detection process. This option may be disabled if automatic semantic type detection is not available for the datasource.
      Note: If automatic detection is not available for a field, the Detect Semantic Type option will be disabled in the Update Semantic Type dialog. In this case, you can only manually assign a semantic type.
    • Create New Semantic Type from Profile: Creates a new custom semantic type with values derived from the asset's profile data. The creation form is pre-populated with name, qualifier, data regex, header regex, base type, priority, and threshold. You can edit all fields before saving and choose between regex based or list based matching. For more information, see the nested task "Create a custom semantic type from profile" below.
    • Manually Assign Semantic Type: Allows you to select a different semantic type from the available list. This option disables automatic semantic type detection for this field.
  6. Select Manually Assign Semantic Type to assign or override the current assignment.
  7. From the dropdown list, search for and select the new semantic type you want to assign. You can type to filter the available semantic types by name or qualifier (for example, 'Email Address'). The list displays the following information for each semantic type:
    • Semantic Type Name: The name of the semantic type.
    • Qualifier: The standardized code or identifier for the semantic type.
    • Priority: The priority level used for detection matching.
    • Base Type: The data type (for example, String, Integer).
    • Source: Whether the semantic type is built-in or custom.
  8. Select the semantic type you want to assign by clicking the radio button next to it.
  9. Click Assign Semantic Type to confirm the assignment.
    Note: After the semantic type is manually assigned, it will not be detected and changed when data is cataloged.

The semantic type is now manually assigned to the field. The system displays a confirmation message indicating that the assignment is complete. The field details page shows the assigned semantic type with an Overridden badge. The manually assigned semantic type will be used for rule detection and data quality scoring, and automatic semantic type detection will not override this manual assignment.

You can track all changes to the semantic type assignment by going to the Change History tab on the field details page. This shows who made the change, what was changed, and when the change occurred.

Create a custom semantic type from profile

Create a custom semantic type directly from an asset's profile data when advanced profiling has detected a regex pattern.

To use the Create New Semantic Type from Profile action, the asset must meet the following conditions:

  • The asset must be advanced-profiled.
  • The profile data must include a generated regex pattern derived from profiling.

If these conditions are not met, the option to create a new semantic type from the profile will not be available.

This action creates a new semantic type using values derived from the asset's profile. The system automatically populates the creation form with key attributes, including name, qualifier, data regex, header regex, base type, priority, and threshold.

Before saving, you can review and modify any of these fields to refine the semantic type definition as needed.

  1. Go to Catalog > Fields.
  2. Click the field name to open the field details page.
  3. Go to the Details or Profile tab.
    The Profile tab displays profiling metrics and analysis for the field, including any advanced profiling patterns.
  4. Verify that a regex pattern is present in the profile data.
    Note: If no regex pattern is shown, advanced profiling has not run or did not detect a pattern. In this case, you cannot use this option for that field.
  5. In the Summary section, locate the Semantic Type field and click the edit icon next to it.
  6. In the Update Semantic Type dialog, click the Create New Semantic Type from Profile action.
    A creation dialog opens with fields pre-populated from the profile data.
  7. Review the pre-populated fields and edit as needed:
    • Name: The asset display name. Change this to a more descriptive name if desired.
    • Qualifier: A unique identifier derived from the asset name. Must match the pattern [A-Z][A-Z0-9_.<>-]*. This field is immutable after creation.
    • Description: You must provide a description (required, maximum 4000 characters).
    • Data regex: The regex pattern pre-filled from the FTA profile. Edit if needed.
    • Header regex: Auto-generated as a case-insensitive match on the column header name. Edit if needed.
    • Base type: Defaults to STRING. Change if the field uses a different data type.
    • Priority: Defaults to 10 (range 0–2500). Lower numbers indicate higher priority during detection.
    • Threshold: Defaults to 90 (range 0–100). Represents the percentage of valid values required for the semantic type to be applied.
  8. Select the Match type:
    • Regex (pattern based): Matches data against a regular expression pattern. This is the default when a regex is present in the profile. You can edit the pre-populated regex or create a new one. Use this for pattern matching (for example, email addresses, phone numbers).
    • List (enumeration based): Matches data against a predefined list of allowed values. When you switch to this type, the content members field is empty and you must manually provide the values (for example, ["ENGLISH", "FRENCH", "SPANISH"]). Profiling does not produce list data, so you are responsible for defining the enumeration.
  9. Optional: Use the Test buttons to validate your regex patterns before submission.

    The form provides two separate Test buttons: one to test the data regex against sample values and one to test the header regex against sample column names. The system uses the Java regex engine to validate patterns. If validation fails, an error message describes the issue. Correct the pattern and test again.

    Note: The system rejects patterns containing anchors (^ or $).
  10. Click Create Semantic Type to save the semantic type.

    If the qualifier already exists in your workspace, the creation fails with a conflict error. When this occurs, rename the qualifier to make it unique and try again.

The semantic type is now created and available in your workspace. It receives higher detection priority in subsequent profiling runs. You can view, edit, or delete the semantic type from the Configuration menu under Catalog > Semantic Types.

You can now assign this semantic type to other fields or use it during data profiling. Follow the steps in the "Assign a semantic type to a field" task above to assign this newly created semantic type to other fields.

Update or remove an existing semantic type

Change or clear the semantic type assignment for a field.

You can update an existing semantic type assignment when you need to change the semantic type that was previously assigned to a field, either automatically or manually. Updating the semantic type allows you to correct misidentified types or assign a more appropriate type.

  1. To update an existing semantic type assignment, follow the same steps as described in the "Assign a semantic type to a field" task above.
    • When you select a new semantic type and click Assign Semantic Type, the new semantic type will replace the existing assignment.
    • The field details page will display the updated semantic type with an Overridden badge.
  2. To remove a semantic type assignment and reset the field to the "Not detected" state, click the edit icon next to the Semantic Type field in the Summary section.
  3. In the Update Semantic Type dialog, select Clear Semantic Type Assignment.
  4. Click Clear to confirm the removal.

The semantic type assignment is now removed. The field returns to the "Not detected" state, and automatic semantic type detection can run again on the next cataloging operation.