Adaptive Feature Classifier

The Adaptive Feature Classifier (AFC) is a content-based classifier that uses the text in a document to classify a document.

The AFC is trained by having it analyze several dozen sample text or XDoc documents per class. It automatically and adaptively determines the important 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 recognition or typing errors may still be used to accurately classify the document. The sample documents are analyzed during Adaptive Feature Classifier training and a classification pattern is automatically created that can be used during production.

Important After changing the properties of your classifiers, or after adding or deleting documents from your training set, you must retrain your project.

To configure the AFC:

  • Enable content classification training for each class that uses this classifier

  • Add training documents from another document subset or add new training documents to your Classification Set

  • Train the project for content classification

Note This classifier supports the following character sets only.
  • Latin-1

  • Greek

  • Russian

  • Armenian

  • Hebrew

  • Arabic

  • Japanese

  • Chinese

  • Korean