Semantic annotation and information extraction are of paramount importance in the field of unstructured information management. Although several sophisticated and indeed complex approaches were proposed, they still have many limitations. In this paper we present a novel ontological paradigm called Description Ontology in which objects and classes (concepts) can be equipped by a set of rules named Concept Descriptors. These rules represent patterns that express how to automatically recognize ontology concepts within unstructured documents. A Description Ontology allows the semantic extraction of information from unstructured document which, in turn, can be stored in structured forms. Moreover, document contents can also be annotated with respect to the recognized concepts. © 2008 IEEE.
Description ontologies
Oro E.Primo
;
2008
Abstract
Semantic annotation and information extraction are of paramount importance in the field of unstructured information management. Although several sophisticated and indeed complex approaches were proposed, they still have many limitations. In this paper we present a novel ontological paradigm called Description Ontology in which objects and classes (concepts) can be equipped by a set of rules named Concept Descriptors. These rules represent patterns that express how to automatically recognize ontology concepts within unstructured documents. A Description Ontology allows the semantic extraction of information from unstructured document which, in turn, can be stored in structured forms. Moreover, document contents can also be annotated with respect to the recognized concepts. © 2008 IEEE.| File | Dimensione | Formato | |
|---|---|---|---|
|
Description_Ontologies.pdf
solo utenti autorizzati
Tipologia:
Versione Editoriale (PDF)
Licenza:
NON PUBBLICO - Accesso privato/ristretto
Dimensione
322.71 kB
Formato
Adobe PDF
|
322.71 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.


