beautypg.com

Document_beforeprocessxdoc, Document_afterclassifyxdoc, Extraction – Kofax Transformation Modules Invoice Pack 1.0 User Manual

Page 11

background image

Configuration

Kofax Transformation Modules Invoice Pack - Configuration Guide

11

Document_BeforeProcessXDoc

This event is triggered before any processing is carried out. The scripted event is
called to check that the incoming document is an image to be OCRed and extracted,
rather than an image with data from an electronic transaction. If it finds that the
required information already exists, it suppresses the OCR and sets a flag to be used
in later script calls.

Document_AfterClassifyXDoc

This event is triggered after the initial document classification. which creates a
reference to the Invoice Pack enhanced logic. The scripted event associated with this
event call will use the default values set within the config.xml file to classify the
document and set the locator keywords to the default language

Kofax Transformation Invoice Packs supports subclasses within Kofax
Transformation Modules.

The specific locator keywords used depend on the Invoice Pack installed – for
example if the UK (en-GB) pack has been installed, then a list of English keywords
are included in the configuration directory along with the config.xml file.

Next, the Invoice Date validation is configured based on the settings in the
config.xml file, after which the OCR confidence values are taken from the config.xml
file and assigned to the fields. The OCR confidence functionality will be covered in
more detail in the extraction section.

Extraction

The extraction fields in the project are populated through a variety of methods. Some
of the fields are modified directly from the script, while others gain their results from
the various format locators and standard evaluators. In some cases, the fields are not
used in the current version of the Invoice Pack, and are populated by simple default
values.

This section will give an overview of each core field, along with how they are
populated and validated. The locators used will be described, and any advanced
logic will be briefly discussed. The OCR confidence functionality, which exists for all
fields, will be detailed at the end of the section. For further information on any of the
locators used, please see the Kofax Transformation Modules documentation.