Create Document Type

Once a project is created and opened, you can create a new Document Type by clicking on the New button and selecting the Using Semi-Structured AI option. A new window opens requesting additional information.

Here are the available options for when creating a new Document Type session. By filling in these values you obtain an estimate of the recommended size of the dataset to be used as a starting point.

Also, by selecting the Out-of-the-box document type, your schema is automatically populated and configured to allow you to benefit from the pre-trained models available in AI Center. This eliminates the need to import predefined schemas manually, accelerates your work, and reduces costly mistakes.

In addition, the Prelabelling endpoint is automatically populated with the appropriate endpoint, making it more convenient for you to use Prelabelling immediately after opening the new Document Type.

Option	Description
Name (Mandatory)	Provide a name for the new document type.
Out-of-the-box document type (Mandatory)	Select one of the pre-trained out-of-the-box document types available in the dropdown list.
Out-of-the-box regular fields (Optional)	Select which of the pre-defined regular fields you would like to create for your schema
Out-of-the-box column fields (Optional)	Select which of the pre-defined column fields you would like to extract from your documents.
Out-of-the-box classification fields (Optional)	Select which of the pre-defined classification fields you would like to extract from your documents.
Custom regular fields (Optional)	Enter the number of additional regular fields you would like to extract from your documents.
Custom column fields (Optional)	Enter the number of additionalcolumn fields you would like to extract from your documents.
Number of languages (Optional)	Enter the number of languages for the documents you need to extract.
Number of layouts (Optional)	Enter the number of layouts for the documents you need to extract.

Note: Selecting a document type generates a recommended number of pages that need to be used for the dataset.