This guide helps you decode the display of the extraction module and understand how to enrich your attributes effectively.
How does the extraction module work?
“Fields” Section
From the "Fields" section on the left side, you can visualize the attributes that the AI model has used to fill in the missing attribute values.
"Attributes" Section
From the "Attributes" section, you can confirm the changes and save them.
There are 2 types of attributes:
- "Select" attributes: where the expected value is proposed in a list.
- "Textual" attributes: where you can complete with text or numbers.
The orange "!" (important) sign appears when the attribute in question has been set as important to complete. If it is not completed you will be alerted but can still proceed to the next step.
Use the filter “Origin” to audit only attributes that the AI model has completed. you can identify these attributes with the robot icon.
There are 3 statuses for the attributes:
- Mandatory represented by a red star, will be blocked if it remains empty.
- Important represented by an orange dot, alerts if the attribute is empty but not blocking .
- Optional is not represented in the application and is non-blocking in all cases.
You can use the filter “Importance” to audit attributes by their level of requirement.
(This setup is made in your project configuration, you can ask the support team to help you change it if needed.)