Adaptive feature classifier, Concept, Set up – Kofax Getting Started with Ascent Xtrata Pro User Manual

Page 103

Advertising
background image

Chapter 3

84

Ascent Xtrata Pro User's Guide

Adaptive Feature Classifier

Concept

The Adaptive Feature Classifier (AFC) is a content-based classifier that uses the text
in a document to identify the class. The AFC is trained by having it analyze several
dozen sample text or XDoc documents per class. It automatically and adaptively
determines the salient features that can be used to define a class. Since the AFC is
fault tolerant, and does not only use words as features, information with OCR or
typing errors may still be used to accurately classify the document. The sample
documents are analyzed during AFC training and a classification pattern is
automatically created that can be used during production.

Set Up

The AFC is automatically inserted into the current project, when sample text or XDoc
files are added to a class and the option “Use for content classification” is selected in
the pop-up menu of the class.

The first time a sample document is added to a class in the project, the message “Do
you want to add text classification support to this project?“ will display. Click Yes to
add the AFC to the current project. For every additional document that is added to
the class, you can decide if it should be used by the AFC. Existing samples can be
removed using the training set viewer for the AFC.

Before the AFC can be tested, it must be trained with the samples. The training step
is required to extract the relevant features from all sample text files and store them in
the project. To train the classifier, select Process | Train Project from the main menu
or click Train Project from the toolbar. A progress bar showing the current status is
displayed while training is performed.

X

To insert a sample document for content classification

1

Select a class in the hierarchy where you want to add the sample text files.

2

Use Windows Explorer or select a Reference Set to open a folder that contains
the sample text files you want to add to the training set.

3

Select the desired file and drag it to the class in the hierarchy.

Advertising