HCONE-merge has been proposed as one of the main functionalities of Human Centered Ontology Engineering Environment (HCONE), developed within an internal project of University of the Aegean, Dept. of Information and Communication Systems Eng., AI-Lab. HCONE-merge makes use of the intended informal meaning of concepts by mapping them to WordNet senses using the Latent Semantic Indexing (LSI) method. Based on these mappings and using the reasoning services of Description Logics, HCONE-merge automatically aligns and then merges ontologies. Since the mapping of concepts to their intended meaning is an essential step of the HCONE-merge approach, we have explored the level of human involvement required for mapping concepts of the source ontologies to their intended meanings. We have proposed a series of methods for ontology mapping (towards merging) with varying degrees of human involvement and evaluate them experimentally. We have shown that we can reach a point where the process of ontology merging can be carried out efficiently with minimum (or non in some cases) human involvement.
The HCONE-merge basic methods place human involvement at the early stages of the merging process. If this involvement leads to capturing the intended meaning of conceptualisations, then the rest is a consistent, error-free merging process, whose results are subject to further human evaluation. The methods proposed show that human involvement is necessary to produce valid mappings between ontologies, however this involvement can be reduced significantly. Major points that differentiate HCONE-merge from other approaches are the following:
On the other hand, the current implementation of the HCONE-merge can not be considered for use for fine-grained domain ontologies: Highly technical terms do not have an entry to WordNet resulting to the poor performance of the method.
In the latest (May, 2005) implementation of our methods, the "approximation of intended meaning of concepts iteratively", no human involvement is needed).Towards automating the HCONE-merge method for merging ontologies, we have proposed a method for aligning the original ontologies with a hidden intermediate ontology in a fully automated way. The alignment is done by mapping ontology concepts to WordNet senses. These senses are supposed to express the human oriented informal intended meanings of ontology concepts. The algorithm proposed is based on previous efforts to approximate similarities between concepts in an iterative way and, as it has been shown, produces mappings that are quite precise.
Current HCONE-merge implementations (in Java) serve as proof-of-concept systems. We are planning a public available version of our latest techniques, incorporating latest technologies, in the near future. A detailed description of the techniques and implementations as well of the experiments conducted so far, can be downloaded from here ("basic" method) and here ("extended" method).
The main set of the ontologies used for our experiments can be found here (in NeoClassic Description Logic). Filenames are denoting the context of the ontology and number of concepts described.
K. Kotis, G. Vouros, K. Stergiou. Towards Automatic Merging of Domain Ontologies: The HCONE-merge approach. Elsevier's Journal of Web Semantics (JWS), vol. 4:1, pp. 60-79, January 2006 . Available on line at (ScienceDirect): http://authors.elsevier.com/sd/article/S1570826805000259
G. Vouros and K. Kotis. Extending HCONE-merge by approximating the intended interpretations of concepts iteratively. 2nd European Semantic Web Conference, Heraklion, Crete May 29 – June 1, 2005. Proceedings, Series: Lecture Notes in Computer Science, Vol. 3532, Asunción Gómez-Pérez, Jérôme Euzenat (Eds.), Springer-Verlag.
K. Kotis, G. Vouros, K. Stergiou. Capturing Semantics towards Automatic Coordination of Domain Ontologies. Artificial Intelligence: Methodology, Systems, and Applications, 11th International Conference, AIMSA 2004, Varna, Bulgaria, September 2-4, 2004. Proceedings, Series: Lecture Notes in Computer Science, Subseries: Lecture Notes in Artificial Intelligence, Vol. 3192, Bussler, Christoph; Fensel, Dieter (Eds.), Springer-Verlag.
K. Kotis, G. Vouros. HCONE approach to Ontology Merging. The Semantic Web: Research and Applications, First European Semantic Web Symposium, ESWS 2004, Heraklion, Crete, Greece, May 10-12, 2004, Proceedings, Series: Lecture Notes in Computer Science, Vol. 3053, Davies, J.; Fensel, D.; Bussler, C.; Studer, R. (Eds.), Springer-Verlag.
Everything (projects, publications, surveys, experiments) about matching ontologies can be found in the new link www.ontologymatching.org
We are currently working on a new approach towards automating the mapping ontologies, integrating the latest technologies available (WordNet 2.0, COCLU lexical matching algorithm, OWL, SMR structural matching algorithm). For more information please read AUTOMS approach to ontology mapping.
K. Kotis at kkot
at aegean dot gr
Copyrights © www.icsd.aegean.gr/ai-lab . All rights reserved. Last updated: 31-12-2005.