An XML-UNL Model for Knowledge-Based Annotation
Jesús Cardeñosa, Carolina Gallardo, Luis Iraola · 2005
Abstract. Efficient document search and description has radically changed with the widespread availability of electronic documents through Internet. Nowadays, efficient information search systems require to go beyond HTMLannotated documents. Complex information extraction tasks require to enrich text with semantic annotations that allow deeper and more detailed content analysis. For that purpose, new labels or annotations need to be defined. In this paper we propose to use UNL, an interlingua defined by the United Nations University, as a language neutral standard content representation in Internet. The use of UNL would open documents to a new dimension of semantic analysis, thus overcoming the limitations of current text-based analysis techniques. 1