Taxonomy-based annotation of XML documents: Application to eLearning resources

Abstract

In this paper we propose an automatic mechanism for annotating XML documents. This mechanism relies on a simple data model whose main features are: (1) a modeling of XML documents as trees composed of elements that are possibly distributed over a network, (2) a composition operator to create new documents from existing ones, and (3) an inference algorithm for automatically deriving the annotation of composite documents from the annotations of their components. We illustrate the features of the model with an application to eLearning resources. We also describe a prototype which allows to create a new document from eLearning fragments collected over the Web, and generates an RDF-based annotation of the document's content. The RDF output can then be used as a support for browsing and querying, by users wishing to create new documents.