The The Phrase Classify Engine classifies documents by comparing the document's OCR result with a list of allowed and forbidden words and phrases. Within certain error limits, the phrases or words must literally occur within the documents.
The engine compares the document's OCR result with the defined lists and determines a phrase weighting.
The engine determines a phrase weight that is used for the class, with the threshold and distance settings of the project, against the weighting of the classification results of other classes and other engines.
This engine is a powerful complement to content classification, since you can define certain keywords that must lead to a specific class. You do not have to define every possible phrase, since the engine does that job.
You can use this engine as the only classification engine. However, this requires the set up and maintenance of a potentially complex set of rules. Typically, you use the Phrase Classify Engine in conjunction with the Brainware Classify Engine.