Manage the field groups, fields, field types, and instructions that define the data an Extract node captures from documents in UiPath Maestro™ Flow.

Prompt instructions are natural-language descriptions that give the model the context it needs to extract the right values from your documents. You can add instructions at several levels of the taxonomy:

Overall extraction instructions - Provide context to the model that is relevant across the taxonomy, including details on the overall extraction task and the documents in the project.

Instructions - Provide context for the model on how to successfully extract data from the document. Iterating on instructions helps to improve predictions.

To add prompt instructions for the entire taxonomy, proceed as follows:

Manage field groups directly from the Taxonomy tab. This allows you to create, edit, move, or delete multiple items before saving your edits. All changes are applied at once when you save them and are saved as a single project version update, instead of generating a new version for each individual change.

To create and configure field groups, proceed as follows:

To define a hierarchy between field groups, create a parent field group and then add one or more child field groups within it. This structure establishes the relationship between them, which means that if no predictions are returned for the parent field group, the child field groups will also return no predictions. The parent field group effectively serves as the primary classification, with its children providing additional levels of detail.

The following example shows field groups and their hierarchies:

To edit field groups, proceed as follows:

To view the change history of a field group, proceed as follows:

Result: The Change history window opens. It lists the versions of the field group, with the field group name and instruction of the selected version. For details on comparing and restoring versions, check Viewing the change history.

You can copy all field definitions of a field group and paste them into one or more other field groups.

To copy field definitions, proceed as follows:

Result: The copied fields are added to the selected field groups.

You can move a field group under another field group, or move a child field group back to the top level.

To move a field group, proceed as follows:

Result: The field group appears under its new parent.

To delete a field group, proceed as follows:

Result: The field group and its fields are removed from the taxonomy.

Manage fields directly from the Taxonomy tab. Create, edit, or delete multiple fields before saving, and all changes are applied at once when you save them.

To create and configure fields, proceed as follows:

You can change the type of a field later. For the effect on existing annotations, refer to the Editing fields section.

To reorder fields within a field group, use the drag handle at the start of the row.

To edit fields, proceed as follows:

To view the change history of a field, proceed as follows:

Result: The Change history window opens. It lists the versions of the field, with the field name, field type, and instruction of the selected version. For details on comparing and restoring versions, check Viewing the change history.

You can copy a field definition and paste it into one or more field groups.

To copy a field definition, proceed as follows:

Result: The copied field is added to the selected field groups.

To delete a field, proceed as follows:

Result: The field and its annotations are removed from the taxonomy.

Deleting a field does not delete the custom field type it uses. To delete the field type, refer to the Deleting field types section.

Field types let you share instructions across all the fields that use them. Manage field types directly from the Field types tab. Like fields and field groups, you can create, edit, or delete multiple field types before saving, and all changes are applied at once when you save them.

To create and configure field types, proceed as follows:

You can also create a field type while you configure a field, by selecting New field type from the Type dropdown list of the field in the Taxonomy tab. The Field types tab opens with a new field type row, and the field uses the new field type.

The following data types are available:

To edit field types, proceed as follows:

To delete a field type, proceed as follows:

Result: The field type is removed from the taxonomy.

You can delete both custom and pre-configured field types. If fields use the field type that you delete, the Delete field type window lists those fields. From the Replace with type dropdown list, select the field type to assign to them. If you select Don't replace, the listed fields have no type, and you must select a new type for each of them in the Taxonomy tab before you can save your changes.

To view the change history of a field or field group, proceed as follows:

I work for the underwriting operations team at an insurance company, and we have hundreds of set policy categories, for example, automotive, home, health, luxury goods, and so on, that we offer to customers, each corresponding to a Type category, for example, Type A, B, C, and so on.

Based on the content of the document, I want to be able to extract and identify the Type category of the policy that needs to be processed.

In this example, there is nowhere in the message that explicitly states that this email pertains to Type E. In the instructions, context is provided for each insurance type to inform the predictions of the model. For example, claims related to luxury goods all belong to the Type E category.

Examples of fields that might need to be set up as Inferred:

To facilitate this request, I may need the existing policy number, name, and the value claimed. These are values that I know need to be explicitly stated in the document itself and extracted into a downstream process.

Figure 1. Example of inferred and exact fields