Tagging for improved semantic interpretation of XML

The development of the Semantic Web depends on agreed and unambiguous knowledge representations, on the availability and accessibility of knowledge, and on retrieval capabilities. The scarce agreement on knowledge representation and the lack of techniques to process semantic structures in web search engines makes it impossible to perform contextualized conceptual retrieval. These limitations imply that users must know beforehand the existence and location of this knowledge to be able to retrieve it. Thus, different ad-hoc knowledge representations and metadata vocabularies, scarcely formalized and agreed on, have been published, hindering the reuse and interoperability. Our proposal has as its main goal the elaboration of different views to improve the representation of semantic documents. This approach facilitates the management and retrieval of heterogeneous semantic schemas by means of a multilevel ontological structure and the alignment with a reference ontology that provides conceptual retrieval and reuse of knowledge.