Set up extraction

One of the main purposes of Kofax Transformation Modules is to extract data from documents. The extracted data is stored in fields. The fields and field extraction definitions are set up in Project Builder and synchronized with Kofax Capture index fields.

Important Kofax Transformation Modules does not perform extraction for unclassified documents. Therefore, it is important to have successful classification.

Kofax Transformation Modules performs extraction by completing two steps.

  1. Locating data on a document

  2. Assigning the results of the location algorithm to one or more fields

These methods for locating items on the documents are implemented in standardized engines called locators.

This two-step process provides reusable locator components and a clear extraction configuration strategy. During production, Kofax Transformation Modules can use several locators to find one field item and can populate several fields from one locator. This ability means that the locator order is extremely important; if necessary, you can rearrange locators.

Kofax Transformation Modules comes with a set of predefined extraction methods. Use these methods to create customized locators. Or, add your own custom locators by providing a script plug-in.

You can define fields at the project level, the field content is extracted at the beginning of classification. The extraction results for these fields may be used to classify a document. For example, classify a document based on a bar code, or perform a language-dependent classification using a Classification Locator, where the locator result is saved to a field at the project level.

If you define fields for a class, subclasses inherit the field definitions. For a subclass, either use the definitions inherited from the base class or change the extraction methods for the fields.

Some locators available in Project Builder are licensed separately for use in production. If you configure a project to use one of those locators, Server does not start unless the required license is available. Any batch based on a project containing unlicensed features is rejected and sent to the Kofax Capture Quality Control module.

You can configure extraction by following these steps:

  1. In the Project Tree, expand your project hierarchy and View the items in the class to configure.

    A list of all class items, including locators and fields, is displayed for the selected class.

  2. Optionally, add and configure a new locator method.
  3. Optionally, modify an existing locator method.
  4. Optionally, add a new field for the selected locator method.
  5. Optionally, modify an existing field for the selected locator method.
  6. Optionally, modify the validation form to accommodate the new class and fields.
  7. Test the locator method with several documents in a document set.
  8. Optionally, generate an Extraction benchmark to compare your test results.
  9. Close your project when you finish testing your changes.