Stage-oe-small.jpg

Inproceedings1028

Aus Aifbportal
Wechseln zu:Navigation, Suche


Bootstrapping Ontology Alignment Methods with APFEL


Bootstrapping Ontology Alignment Methods with APFEL



Published: 2005 November
Herausgeber: Y. Gil, E. Motta, V. R. Benjamins, M. A. Musen
Buchtitel: Proceedings of the 4th International Semantic Web Conference, ISWC 2005, Galway, Ireland, November 6-10, 2005.
Ausgabe: 3729
Reihe: LNCS
Seiten: 186-200
Verlag: Springer

Referierte Veröffentlichung
Note: Nominated for best paper award!

BibTeX

Kurzfassung
Ontology alignment is a prerequisite in order to allow for interoperation between different ontologies and many alignment strategies have been proposed to facilitate the alignment task by (semi-)automatic means. Due to the complexity of the alignment task, manually defined methods for (semi-)automatic alignment rarely constitute an optimal configuration of substrategies from which they have been built. In fact, scrutinizing current ontology alignment methods, one may recognize that most are not optimized for given ontologies. Some few include machine learning for automating the task, but their optimization by machine learning means is mostly restricted to the extensional definition of ontology concepts. With APFEL (Alignment Process Feature Estimation and Learning) we present a machine learning approach that explores the user validation of initial alignments for optimizing alignment methods. The methods are based on extensional and intensional ontology definitions. Core to APFEL is the idea of a generic alignment process, the steps of which may be represented explicitly. APFEL then generates new hypotheses for what might be useful features and similarity assessments and weights them by machine learning approaches. APFEL compares favorably in our experiments to competing approaches.

ISBN: 3-540-29754-5
ISSN: 0302-9743
VG Wort-Seiten: 27
Download: Media:2005_1028_Ehrig_Bootstrapping_O_1.pdf
Weitere Informationen unter: Link

Projekt

SEKTSTI2SWAPKnowledgeWeb



Forschungsgruppe

Wissensmanagement


Forschungsgebiet

Maschinelles Lernen, Semantic Web