Classification Locator sample
This
sample project was created to be
used as an
embedded
project. This means that you can configure a
Classification Locator and reference this sample project rather than create a classification project
from scratch. The sample project is pre-configured for layout or content classification, either instructions or
Adaptive Feature Classifier and has a subdirectory structure that can be used to classify a document
by its language. When processing documents, string values describing the languages are subsequently returned.
After you have extracted the CLSLoc_Language sample zip file, you can find the project file and another file needed for
content classification. The multi-lingual documents used for training are not part of this sample project, because all training
information is saved to the
Language_Content Classifier.dat
file.
When setting up your extraction project, a project that does not have a
Classification Set yet is sufficient. Set up this project with a default classification result (1 base
class), define 1 field and assign it to a
Classification Locator. In the properties of the
Classification Locator, reference the
CLS_loc_language
sample project and leave the rest on default. Now feed your project with documents in various
languages: the configured field contains the language as estimated by the Classification Locator.