Azure Document Intelligence AI Action is a specific implementation of Document AI that uses Microsoft Azure’s Document Intelligence service (formerly Form Recognizer) for document processing. This action type provides specialized prebuilt models optimized for common document types such as invoices, receipts, contracts, and layouts.
Several fields must be specified to configure an Azure Document Intelligence AI Action:
Additionally, model-specific optional fields are:
- Set Properties (Model-Specific)
- Create Object(s) (Model-Specific)
- Set File(s) (Prebuilt-Read)
- Pages
- Confidence Threshold
Below are descriptions of all Azure Document Intelligence AI Action-related fields and their respective values.
Action Type #
Azure AI Action Type
Required field
The Action Type determines the operation that will be performed on the document using Azure Document Intelligence models. Choose from the following options:
- Extract (default): Choose this option to extract structured data from documents using Azure’s prebuilt models. The Extraction Model field becomes available, along with model-specific configuration options for properties, objects, and files.
- Classify: Choose this option to automatically classify documents based on their content. The Classifier Model field becomes available.
Model Selection #
The model selection varies based on the Action Type. Azure Document Intelligence provides specialized prebuilt models, each optimized for specific document types.
Extraction Model
Azure AI Action Type: Extract > Extraction Model
Displayed only when Action Type is set to Extract.
Default value: prebuilt-read
Specifies the Azure Document Intelligence prebuilt model to be used for data extraction operations. Select from the following available models:
prebuilt-read (default) #
General-purpose optical character recognition (OCR) model that extracts text from documents. Ideal for:
- General text extraction
- Multi-language documents
- Scanned documents
- Images with text
When selected, the Set File(s) from prebuilt-read configuration becomes available.
prebuilt-layout #
Extracts text, tables, selection marks, and document structure. Ideal for:
- Forms with complex layouts
- Documents with tables
- Structured documents requiring layout preservation
- Documents with checkboxes or selection marks
When selected, the Set Properties prebuilt-layout and Create Object(s) from prebuilt-layout configurations become available.
prebuilt-invoice #
Specialized model for invoice processing that extracts key invoice fields. Ideal for:
- Vendor invoices
- Purchase invoices
- Service invoices
- Commercial documents
Automatically extracts fields such as invoice number, date, vendor information, line items, totals, and tax information.
When selected, the Set Properties prebuilt-invoice and Create Object(s) from prebuilt-invoice configurations become available.
prebuilt-receipt #
Specialized model for receipt processing that extracts key receipt fields. Ideal for:
- Purchase receipts
- Sales receipts
- Expense receipts
- Point-of-sale documents
Automatically extracts fields such as merchant name, transaction date, items, amounts, and tax information.
When selected, the Set Properties prebuilt-receipt and Create Object(s) from prebuilt-receipt configurations become available.
prebuilt-contract #
Specialized model for contract processing that extracts key contract fields. Ideal for:
- Legal contracts
- Service agreements
- Business contracts
- Terms and conditions documents
Automatically extracts fields such as parties involved, contract dates, clauses, terms, and obligations.
When selected, the Set Properties prebuilt-contract and Create Object(s) from prebuilt-contract configurations become available.
Classifier Model
Azure AI Action Type: Classify > Classifier Model
Displayed only when Action Type is set to Classify.
Specifies the Azure Document Intelligence classifier model to be used for document classification operations. Select from the list of available custom classifier models configured in your Azure Document Intelligence resource.
Note: Azure Document Intelligence classifiers require training with your specific document types before use.
Set Properties (Model-Specific) #
Displayed only when Action Type is set to Extract and a model other than prebuilt-read is selected.
The Set Properties configuration is model-specific and varies depending on the selected extraction model. Each model exposes different fields that can be extracted from documents.
Set Properties prebuilt-layout #
Azure AI Action Type: Extract > Extraction Model: prebuilt-layout > Set Properties prebuilt-layout
Displayed only when Extraction Model is set to prebuilt-layout.
Defines properties that will be populated with data extracted from the document using the prebuilt-layout model. This model can extract:
- Text content from specific regions
- Table data
- Selection marks (checkboxes, radio buttons)
- Document structure information
- Expand the Set Properties prebuilt-layout section.
- Click on Add Property Mapping.
- Configure the property mapping for layout-specific fields.
Set Properties prebuilt-invoice #
Azure AI Action Type: Extract > Extraction Model: prebuilt-invoice > Set Properties prebuilt-invoice
Displayed only when Extraction Model is set to prebuilt-invoice.
Defines properties that will be populated with invoice data extracted by the prebuilt-invoice model. Common invoice fields include:
- Invoice ID / Number
- Invoice Date
- Due Date
- Vendor Name
- Vendor Address
- Customer Name
- Customer Address
- Subtotal
- Tax Amount
- Total Amount
- Payment Terms
- Purchase Order Number
- Expand the Set Properties prebuilt-invoice section.
- Click on Add Property Mapping.
- Select the invoice field to extract and the target M-Files property.
Set Properties prebuilt-receipt #
Azure AI Action Type: Extract > Extraction Model: prebuilt-receipt > Set Properties prebuilt-receipt
Displayed only when Extraction Model is set to prebuilt-receipt.
Defines properties that will be populated with receipt data extracted by the prebuilt-receipt model. Common receipt fields include:
- Merchant Name
- Merchant Address
- Merchant Phone Number
- Transaction Date
- Transaction Time
- Receipt Total
- Subtotal
- Tax Amount
- Tip Amount
- Items (individual line items)
- Expand the Set Properties prebuilt-receipt section.
- Click on Add Property Mapping.
- Select the receipt field to extract and the target M-Files property.
Set Properties prebuilt-contract #
Azure AI Action Type: Extract > Extraction Model: prebuilt-contract > Set Properties prebuilt-contract
Displayed only when Extraction Model is set to prebuilt-contract.
Defines properties that will be populated with contract data extracted by the prebuilt-contract model. Common contract fields include:
- Contract ID
- Contract Date
- Effective Date
- Expiration Date
- Party Names
- Party Addresses
- Contract Type
- Terms and Conditions
- Clauses
- Obligations
- Signatures
- Expand the Set Properties prebuilt-contract section.
- Click on Add Property Mapping.
- Select the contract field to extract and the target M-Files property.
Create Object(s) (Model-Specific) #
Displayed only when Action Type is set to Extract and applicable models are selected.
The Create Object(s) configuration is model-specific and allows automatic creation of related M-Files objects based on extracted data. This is particularly useful for creating line items, related records, or subsidiary documents.
Create Object(s) from prebuilt-layout #
Azure AI Action Type: Extract > Extraction Model: prebuilt-layout > Create Object(s) from prebuilt-layout
Displayed only when Extraction Model is set to prebuilt-layout.
Creates new M-Files objects based on structured data extracted by the layout model, such as table rows or repeated form sections.
- Expand the Create Object(s) from prebuilt-layout section.
- Click on Add New Object Configuration.
- Define the object type, class, and property mappings.
Create Object(s) from prebuilt-invoice #
Azure AI Action Type: Extract > Extraction Model: prebuilt-invoice > Create Object(s) from prebuilt-invoice
Displayed only when Extraction Model is set to prebuilt-invoice.
Creates new M-Files objects based on invoice data, typically for invoice line items. Each line item can be created as a separate object with properties such as:
- Item Description
- Quantity
- Unit Price
- Line Amount
- Tax
- Product Code
- Expand the Create Object(s) from prebuilt-invoice section.
- Click on Add New Object Configuration.
- Define the object type, class, and property mappings for line items.
Create Object(s) from prebuilt-receipt #
Azure AI Action Type: Extract > Extraction Model: prebuilt-receipt > Create Object(s) from prebuilt-receipt
Displayed only when Extraction Model is set to prebuilt-receipt.
Creates new M-Files objects based on receipt data, typically for receipt line items. Each item can be created as a separate object with properties such as:
- Item Name
- Quantity
- Price
- Total Price
- Expand the Create Object(s) from prebuilt-receipt section.
- Click on Add New Object Configuration.
- Define the object type, class, and property mappings for receipt items.
Create Object(s) from prebuilt-contract #
Azure AI Action Type: Extract > Extraction Model: prebuilt-contract > Create Object(s) from prebuilt-contract
Displayed only when Extraction Model is set to prebuilt-contract.
Creates new M-Files objects based on contract data, such as:
- Contract parties (organizations or individuals)
- Contract clauses or sections
- Related obligations or milestones
- Linked documents referenced in the contract
- Expand the Create Object(s) from prebuilt-contract section.
- Click on Add New Object Configuration.
- Define the object type, class, and property mappings for contract-related objects.
Set File(s) (Prebuilt-Read) #
Azure AI Action Type: Extract > Extraction Model: prebuilt-read > Set File(s) from prebuilt-read
Displayed only when Extraction Model is set to prebuilt-read and Action Type is not Classify.
Defines files that will be created or modified based on the text extraction from the prebuilt-read model. This can include:
- Extracted text files
- Processed document versions
- OCR output files
- Searchable PDF versions
- Expand the Set File(s) from prebuilt-read section.
- Click on Add File Configuration.
- Define the file source, format, and destination.
Note: This configuration is unique to the prebuilt-read model and is not available for other prebuilt models.
Pages #
Specifies which pages of the document will be processed by the Azure Document Intelligence model.
Format: Enter a comma-separated list of page numbers and/or page ranges.
Examples:
1– Process only the first page1,3,5– Process pages 1, 3, and 51-5– Process pages 1 through 51,3-5,7– Process page 1, pages 3 through 5, and page 7
Default behavior: If left empty (""), all pages in the document will be processed.
Performance consideration: Processing fewer pages improves performance and reduces Azure consumption costs.
Confidence Threshold #
Azure AI Action Type: Classify > Confidence Threshold
Displayed only when Action Type is set to Classify.
Specifies the minimum confidence level required for the classifier to apply a classification label to a document.
Range: 0.0 to 1.0
Default value: 0.8 (80% confidence)
Behavior: If the classifier’s confidence score falls below this threshold, the classification will not be automatically applied. This prevents uncertain or incorrect classifications from being set on documents.
