Ontology Based Semantic Similarity Comparison of Documents
V. Oleshchuk, und A. Pedersen. DEXA '03: Proceedings of the 14th International Workshop on Database and Expert Systems Applications, Seite 735. Washington, DC, USA, IEEE Computer Society, (2003)
Zusammenfassung
In this work we consider ontologies as knowledgestructures that specify terms, their properties and relationsamong them to enable knowledge extraction fromtexts. We represent ontologies using a graph-basedmodel that reflect semantic relationship between conceptsand apply them to text analysis and comparison.Instead of raw document comparison we compare documentfootprint enhanced with concepts from the ontology(using different enhancement algorithms). Theresult of this process may be that documents not similarprior to the enhancement become similar (semanticallyon some abstraction level) after the enhancement.This is because the enhancement process may introducein the document footprint abstract concepts from theontology. Using the ontology we can enhance the foot-printsby adding concepts that are not present in theoriginal document. We may use synonyms for a horizontalexpansion and broader terms/superclasses/typesin a vertical expansion or both for that matter.
Beschreibung
Ontology Based Semantic Similarity Comparison of Documents
%0 Conference Paper
%1 Oleshchuk03ontologySemantics
%A Oleshchuk, Vladimir
%A Pedersen, Asle
%B DEXA '03: Proceedings of the 14th International Workshop on Database and Expert Systems Applications
%C Washington, DC, USA
%D 2003
%I IEEE Computer Society
%K 03 Oleshchuk documents ontology semantics
%P 735
%T Ontology Based Semantic Similarity Comparison of Documents
%U http://portal.acm.org/citation.cfm?id=942991
%X In this work we consider ontologies as knowledgestructures that specify terms, their properties and relationsamong them to enable knowledge extraction fromtexts. We represent ontologies using a graph-basedmodel that reflect semantic relationship between conceptsand apply them to text analysis and comparison.Instead of raw document comparison we compare documentfootprint enhanced with concepts from the ontology(using different enhancement algorithms). Theresult of this process may be that documents not similarprior to the enhancement become similar (semanticallyon some abstraction level) after the enhancement.This is because the enhancement process may introducein the document footprint abstract concepts from theontology. Using the ontology we can enhance the foot-printsby adding concepts that are not present in theoriginal document. We may use synonyms for a horizontalexpansion and broader terms/superclasses/typesin a vertical expansion or both for that matter.
%@ 0-7695-1993-8
@inproceedings{Oleshchuk03ontologySemantics,
abstract = {In this work we consider ontologies as knowledgestructures that specify terms, their properties and relationsamong them to enable knowledge extraction fromtexts. We represent ontologies using a graph-basedmodel that reflect semantic relationship between conceptsand apply them to text analysis and comparison.Instead of raw document comparison we compare documentfootprint enhanced with concepts from the ontology(using different enhancement algorithms). Theresult of this process may be that documents not similarprior to the enhancement become similar (semanticallyon some abstraction level) after the enhancement.This is because the enhancement process may introducein the document footprint abstract concepts from theontology. Using the ontology we can enhance the foot-printsby adding concepts that are not present in theoriginal document. We may use synonyms for a horizontalexpansion and broader terms/superclasses/typesin a vertical expansion or both for that matter.},
added-at = {2009-11-30T04:17:47.000+0100},
address = {Washington, DC, USA},
author = {Oleshchuk, Vladimir and Pedersen, Asle},
biburl = {https://www.bibsonomy.org/bibtex/204ec138d1de9aba2e8ffbd8a0d00eedc/lee_peck},
booktitle = {DEXA '03: Proceedings of the 14th International Workshop on Database and Expert Systems Applications},
description = {Ontology Based Semantic Similarity Comparison of Documents},
interhash = {b6e101c059e6fe825b4134e129347477},
intrahash = {04ec138d1de9aba2e8ffbd8a0d00eedc},
isbn = {0-7695-1993-8},
keywords = {03 Oleshchuk documents ontology semantics},
pages = 735,
publisher = {IEEE Computer Society},
timestamp = {2009-11-30T04:17:47.000+0100},
title = {Ontology Based Semantic Similarity Comparison of Documents},
url = {http://portal.acm.org/citation.cfm?id=942991},
year = 2003
}