introduction •
ontology •
applications •
publications •
team •
further information
>link to the SeReS II follower project (2008-2010)
Geospatial web services will play a major role in the emerging Semantic Web whose vision is to make web contents accessible to
both human and non-human agents. Using and sharing information in such
environments requires semantic interoperability, which involves capturing
the semantics of applications through ontologies and negotiating between
them.
In analogy to spatial reference systems that enable on-the-fly transformations
of geospatial information between different coordinate systems, we propose
the notion of semantic reference systems, to enable semantic translation
between geospatial information resources, such as databases and web services.
The project Semantic Reference Systems (SeReS) started out with the general
research question of how to enable geospatial information providers to semantically
annotate their contents and how to enable information requesters to discover
suitable information sources for the task at hand. The project's goal is
to develop a theory of semantic reference systems and to provide, as proof
of concept, a prototypical application. The approach incorporates theories
from cognitive science, namely conceptual spaces and image schemas, into
ontology engineering.
We follow the hypothesis that an ontology-based semantic reference system
allows for comparing the conceptualizations underlying information sources.
In a spatial reference system, the distance between two objects can be determined
and transformations of coordinates can be calculated. In analogy, a semantic
reference system allows to determine the "distance" between two
concepts and semantic translations between "locations" of concepts
in semantic reference systems. Enabling information providers and requesters
to compare, align or transform their conceptualizations is an essential
requirement for achieving semantic interoperability.
After a successful completing of the first phase of the project, we applied successfully for a continuation of the project. The start of the second phase is currently under preparation. In this phase we focus on developing a theory for spatial relational qualities.
overview
overview
three level ontology architecture
(extended) foundation ontology
Foundation ontologies provide a rigorous formal semantics for the top-level
notions. They are intended to serve as conceptual foundation for domain ontologies,
which in turn provide specialized views on a certain domain. The semantics
of the vocabulary used in domain ontologies are specified using top-level
concepts.
The foundation ontology DOLCE
(Descriptive Ontology for Linguistic and Cognitive Engineering) is
used in the SeReS ontology architecture. We analyze to which extent the current
version of DOLCE is suitable to account for observations & measurements. In this sense, we develop an extension to the foundation ontology DOLCE. The general structure needed to communicate about observation and measurement is domain independent.
OWL-based Version 1.0 (June 2007):
extended foundation ontology accounting for observation & measurements
Formal characterization based on first order logics. (coming soon)
ontology
domain ontologies
- OGC Observation & Measurements
We provide a mapping of the current O&M model into the ontology web
language
OWL. In a first step, the current O&M model was mapped as directly as
the different
representation formats allow. This resulted in a first draft O&M ontology.
This
ontology formed the basis to evaluate the O&M model from an ontology
engineering
point of view. The analysis provides new insights into the conceptual model
of O&M
and leads towards a semantic framework for semantically annotating the central
terms
in O&M.
Discussion paper on turning the
O&M model into an OWL-based ontology.
raw O&M ontology
modified O&M ontology
- Gewaesserstrukturgueteklassifizierung LAWA
The German classification schema for the
ecological assessment of watercourse structure (Gewässerstrukturgütekartierung)
[28] provides an excellent opportunity to investigate the kinds of basic
qualities and their possible combinations into complex qualities. The second
use case investigated how to turn the text-based classification schema into
a DOLCE-aligned domain ontology. The goal was to understand the structure
of an ontology of physical qualities. This ontology of qualities assists
an information provider in finding an ontologically consistent description
of the qualities for which her data set provides observation results, e.g.
water temperatures or river depths. Since the classification schema is part
of the German realization of the European water framework directive, the
ontological analysis of its concepts supports the use of water quality data
in spatial data infrastructures like INSPIRE [29].
ontology
application ontologies
In the developed application ontologies, we take some of the qualities specified in the gwsg domain ontology and specify complete semantic reference spaces for those.
For example, the general domain qualities "river-segment-depth" and "profile-depth" are fully specified.
Version 1.0 (June 2007):
SeReS application ontology for ecological assessment of watercourse structure
(Requires the import of version 1.0 of the extended foundational ontology. See above)
ontology
boathouses & houseboats
In this use case we investigate the possibilities of using Haskell as ontology
language. Focus is on capturing the semantics of houseboats and boathouses
via the affordances boats and houses offer.
Ontologies written in Haskell
(Please get in contact for more details!)
ontology
image schema ontology
currently under revision. Please get in contact for more details!
This ontology contains symbols for image schema. Since the
currently available ontology editors are string-based, the image schemas
are represented with terms. This may be misleading in indicating that
an image schema can be represented by a term (symbol)
ontology
prototype
SeRes VIS-Tool
We developed a prototype for visually comparing semantic reference spaces.
(due to building renovation, server is currently down (Oct. '07). Please get in contact for any questions.
Start application
publications
Kuhn, W. (2004). Elements of a Computational Theory of Location. AGILE,
Heraklion, Crete.
M. Raubal (2004) Formalizing Conceptual Spaces. in: A. Varzi and L.
Vieu (Eds.), Formal Ontology in Information Systems, Proceedings of the Third
International Conference (FOIS 2004). Frontiers in Artificial Intelligence
and Applications 114, pp. 153-164, IOS Press, Amsterdam, NL.
Probst, F. and Lutz, M. (2004): Giving Meaning to GI Web Service Descriptions
(Extended Abstract). 7th Conference on Geographic Information Science
(AGILE 2004), Heraklion, Greece. pdf
Probst, F., F. R. Gibotti, A. M. Prazos Morantes, M. A. Esbri, M. B. B. de
Barros Filho, M. Gutierrez and W. Kuhn (2004). Connecting ISO and OGC Standards
to the Semantic Web. Third International Conference on Geographic Information
Science, Adelphi, MD, USA. pdf
M. Raubal (2005) Mappings For Cognitive Semantic Interoperability.
in: F. Toppen and M. Painho (Eds.), AGILE 2005 - 8th Conference on Geographic
Information Science. pp. 291-296, Instituto Geografico Portugues (IGP), Lisboa,
Portugal.
Klien, E. and F. Probst (2005). Requirements for Geospatial Ontology Engineering.
8th Conference on Geographic Information Science (AGILE 2005). Estoril, Portugal. pdf
Probst, F. (2005). Position Paper: Investigating the Applicability of DOLCE
as Foundation for Service Descriptions. W3C Workshop on Frameworks for
Semantics in Web Services, Innsbruck, http://www.w3.org/2005/04/FSWS/accepted-papers.html.
W. Kuhn (2005). Geospatial Semantics: Why, of What, and How? Journal
on Data Semantics, Special Issue on Semantic-based Geographical Information
Systems, Lecture Notes in Computer Science, 3534: 1-24.
Probst, F. (2006). An Ontological Analysis of Observations and Measurements.
4th. International Conference on Geographic Information Science (GIScience)
(accepted for publication), Münster, Germany. pdf
Probst, F. and M. Espeter (2006). Spatial Dimensionality as Classification
Criterion for Qualities.FOIS'06 International Conference
on Formal Ontology in Information Systems. pdf
overview
overview
Please contact: Florian Probst
Muenster
Semantic Interoperability Lab (MUSIL)
Institute
for Geoinformatics
overview