The example, on this page, features four classifiers – Naïve Bayesian with tokens feature space, Naïve Bayesian with lemmas feature space, Relative entropy with noun phrases feature space and Class-featured centroid with noun phrases feature space. The classification module registers these algorithms as OSGI services, according to the configuration settings. The features of a new document are extracted by the LPC framework and are passed to the corresponding classifier. Each of the classifiers uses its model to predict a set of labels. Finally, all results are combined and the classification is displayed to the user.
The following diagrams depict the sequence of actions and main participants involved in the building and using of a classification model.
ATLAS (Applied Technology for Language-Aided CMS) is a project funded by the European Commission under the CIP ICT Policy Support Programme.