Validate extraction results to ensure that all information is valid. You can also mark documents for online learning.

Administrators set up extraction fields to extract specific document data and configure the validation layout in the Document tab. Fields marked with warning icons contain errors and must be addressed.

Fields can be invalid for several reasons, including the following:

• Poor OCR results: Some letters and values are not recognized as expected. Manually type the correct value.

• An unconfident or incorrect result: When extraction is configured by your administrator, they may specify a minimum confidence that all fields must meet in order to be valid. If a field is unconfident, the minimum threshold is not met. Alternatively, the extracted result may not meet the configured criteria, such as being a valid data type, or being within a numerical range. In either case, you will need to validate the result.

• A value could not be found on the document: When extraction is configured by your administrator, they may specify that certain fields must be extracted in order for a document of that class to be valid. If extraction fails to return a result, type the valid field value.

• Incorrect classification of the document: Extraction is dependent on the class assigned to a document. As a result, if the document is incorrectly classified, field values may be incorrect. In this case, you will need to assign the correct class and then type the valid field values.

By default, you cannot complete a document that contains invalid fields. However the administrator can override this setting so documents with invalid fields can be marked as complete.

If you cancel the validation process, Kofax TotalAgility saves the invalid fields and any changes.