Kofax Getting Started with Ascent Xtrata Pro - Manual (Page 167)

Chapter 4
148 Ascent Xtrata Pro User's Guide
Classification Locator
The following sections describe the concept of the Classification Locator and show
how to add and set up the locator.
Concept
The Classification Locator uses the classification scheme defined in a secondary
external Ascent Xtrata Pro classification project to provide additional classification
results for a document as field values. Only the classification scheme is used from the
external project.
This offers new possibilities for data extraction. The Classification Locator is useful
whenever you want to add additional information to a document that can be
obtained by additional classification steps. Since the additional classification steps
are normally independent of the main classification in the main project, an externally
defined and trained project is used. The Classification Locator gives access to multi-
view classification that sees the document from different aspects and multi-topic
classification that returns more than one classification result for a document or even
by line of text.
Example use cases for the Classification Locator are listed below:
• Language (simple use-case): If you receive multi-language correspondence,
you can design a project that classifies documents according to their language.
A sample project for this case is provided with Ascent Xtrata Pro and can be
found in the Samples directory. The locator’s result will then be the name of
the class returned from the language classification project (for example,
German or Danish).
• Text body extraction (intermediate use-case): Define a project that classifies
text into two classes using instructions based on salutations (“Dear Miss X”
etc.) and closings (“Yours sincerely” etc.). Save the project and assign it to a
Classification Locator in your current project. Set up the locator to use line-by-
line classification. The result will return alternatives from text lines that match
these instructions in the document. Then, use the Script Locator (see below) to
further evaluate these alternatives. The salutation and closing should be listed
among the best ones. In the script, use the coordinates of those alternatives to
extract the text body of your letter, which is assumed to be the part between
the salutation and closing.
• Product groups (advanced use-case): Since the Classification Locator can be
configured to work line-by-line, you can set up a project that classifies line