Abstract
Traditionally National Mapping Agencies (NMAs) produce maps at different scales. Since automatic generalisation is not yet
possible (it will at least take years) NMAs have to maintain their data sets in a multiple representation environment. Important
principles of this environment are 1) the data sets at the different scales are semantically integrated, 2) objects in the different scales
representing the same real world objects are linked and 3) updates in the largest scale data set are automatically propagated to the
smaller scales. This paper describes the considerations for semantic data integration of the topographic database of the Dutch NMA
(TD Kadaster). In collaboration with ITC a generic data model is being designed in which the data models of the different scales are
integrated. This model supports topographic data sets at a scale range from 1:10k to 1:500k. The results of this data model
integration that were obtained so far are reported in this paper. The difficulties that are met are related to using existing data sets in
the data integration process. For example the classifications of the different data sets (i.e. map legends) are not synchronised, since
this was not needed when producing paper maps. Also the generalisation rules that are verbally described in cookbooks to support
the interactive work of cartographers seem not to be in accordance with the classification schemas of the existing data sets. Apart for
internal processes, semantic data integration is also required for the exchange of geo information with other geo information
suppliers. This paper describes how semantic data integration of Dutch topographic data with the outside national world is assured
by embedding the topographic data model in the Dutch standardised data model for geo information. The embedding is modelled
using Unified Modelling Language. The model is limited in some aspects since GML is used as exchange standard. Therefore this
paper considers the possibilities of a new technology in the semantic data integration process of different data sets: machine
ontology using the Web Ontology Language (OWL).
possible (it will at least take years) NMAs have to maintain their data sets in a multiple representation environment. Important
principles of this environment are 1) the data sets at the different scales are semantically integrated, 2) objects in the different scales
representing the same real world objects are linked and 3) updates in the largest scale data set are automatically propagated to the
smaller scales. This paper describes the considerations for semantic data integration of the topographic database of the Dutch NMA
(TD Kadaster). In collaboration with ITC a generic data model is being designed in which the data models of the different scales are
integrated. This model supports topographic data sets at a scale range from 1:10k to 1:500k. The results of this data model
integration that were obtained so far are reported in this paper. The difficulties that are met are related to using existing data sets in
the data integration process. For example the classifications of the different data sets (i.e. map legends) are not synchronised, since
this was not needed when producing paper maps. Also the generalisation rules that are verbally described in cookbooks to support
the interactive work of cartographers seem not to be in accordance with the classification schemas of the existing data sets. Apart for
internal processes, semantic data integration is also required for the exchange of geo information with other geo information
suppliers. This paper describes how semantic data integration of Dutch topographic data with the outside national world is assured
by embedding the topographic data model in the Dutch standardised data model for geo information. The embedding is modelled
using Unified Modelling Language. The model is limited in some aspects since GML is used as exchange standard. Therefore this
paper considers the possibilities of a new technology in the semantic data integration process of different data sets: machine
ontology using the Web Ontology Language (OWL).
| Original language | English |
|---|---|
| Title of host publication | Proceedings of Joint ISPRS working groups II/3 and II/6 workshop on multiple representation and interoperability of spatial data |
| Subtitle of host publication | February 22-24, 2006 Hannover, Germany |
| Publisher | International Society for Photogrammetry and Remote Sensing (ISPRS) |
| Number of pages | 8 |
| Publication status | Published - 2006 |
| Event | Joint ISPRS working groups II/3 and II/6 workshop on multiple representation and interoperability of spatial data - Hannover, Germany Duration: 22 Feb 2006 → 24 Feb 2006 |
Workshop
| Workshop | Joint ISPRS working groups II/3 and II/6 workshop on multiple representation and interoperability of spatial data |
|---|---|
| Country/Territory | Germany |
| City | Hannover |
| Period | 22/02/06 → 24/02/06 |
Keywords
- GIP
- ADLIB-ART-1316
Fingerprint
Dive into the research topics of 'Semantic data integration in a multiple representation environment'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver