Creating an intelligent structure model using custom AI engine

Create an intelligent structure model based on input that represents the data that you expect the model to parse at run time.

Before you use the custom AI engine to create an intelligent structure model, be sure to complete the following prerequisites:

•Create a custom AI connection. For example, a Large Language Model connection.

Note: You can only use a Large Language Model connection of type chat.

For more information about how to create a Large Language Model connection, see the Large Language Model connector documentation.

•Enable Python Integration service for the Secure Agent group. To do this, perform the following steps:

1In Administrator, select Runtime Environments.
2Right-click the Secure Agent group and select Enable or Disable Services and Connectors.

A dialog box listing all the services and connectors for the Secure Agent group is displayed.

3On the Services tab, enable the Python Integration service.
4Click OK.

After you complete the prerequisites, perform the following steps to create an intelligent structure model using the custom AI engine:

1Click New > Components > Intelligent Structure Model, and then click Create.

2On the Intelligent Structure Model page, enter a name for the intelligent structure model.

The name can contain alphanumeric characters and underscores.

3Navigate to the project and folder where you want to save the model, or use the default location.

You can change the name or location after you save the intelligent structure model using the Explore page.

4Under AI-powered Model, select Custom AI Engine.

You can use the custom AI engine to process images, emails, and text documents. You can extract data from the following input types when you use the custom AI engine:

- Contractual information
- Financial documents
- Invoice copies
- Generic key-value pairs

5In Custom AI Connection, select an existing Large Language Model connection, or click New Connection to create a new Large Language Model connection.

6Based on the type of input, first choose whether to base the model on a file sampling or on the entire file. Select the file, and then click Discover Structure.

After you click Discover Structure, Intelligent Structure Discovery deciphers the data in the input and discovers the patterns expressed in the data. The following image shows an example of discovered structure on the Visual Model tab: The Visual Model tab shows an example of a discovered structure with a hierarchy of nodes.

The Visual Model tab shows an example of a discovered structure with a hierarchy of nodes.

Intelligent Structure Discovery creates nodes with unique names. If Intelligent Structure Discovery discovers instances of the same type of data, it adds a number suffix to the node names. For example, if the input contains timestamps in two tables, Intelligent Structure Discovery names them timestamp1 and timestamp2.

7You can refine the structure so that when you use the model in production, the output meets your requirements. For more information, see Refining intelligent structure models.

8Click Save.

Intelligent Structure Discovery generates an intelligent structure model and saves the model in the selected location.