Unsupervised Web-based Automatic Annotation

9 years 12 months ago
Unsupervised Web-based Automatic Annotation
The success of the Semantic Web depends both on the definition of ontologies used to represent the knowledge as on the annotations performed of the web contents. As manual approaches have low scalability, there is a need of tools capable to generate all this knowledge in an automatic and reliable way. In this paper is presented a complete algorithm to annotate web contents in an automatic and unsupervised manner. It is structured in a three-stepped procedure, based on the usage of several concept similarity measures and linguistic patterns. It is able to detect the entities to annotate, the candidate classes of these entities and, finally, associate them with the classes of an ontology. Some prospective results are presented. Keywords. Semantic web, automatic annotation, ontologies
Miquel Millan, David Sánchez, Antonio Moren
Added 30 Oct 2010
Updated 30 Oct 2010
Type Conference
Year 2008
Authors Miquel Millan, David Sánchez, Antonio Moreno
Comments (0)