Automated spatial data harmonisation

Research project objectives

The objective of the project is a research on the semantic interoperability concept and possibilities of automated spatial data harmonization and integration from National Geodetic and Cartographic Resource with the use of Semantic Web technologies.

INSPIRE defines interoperability as the possibility to combine spatial data sets and services from different sources in a consistent way as well as providing access to it through network services. According to implementation rules, maintaining interoperability of both data sets and services is important task. Providing syntactic interoperability based on well-known data formats and communication protocols is relatively simple, unlike providing semantic interoperability, which can be nontrivial task due to multiple possible data interpretation. Apart from being syntactically interoperable, data sets from two different services can have different meanings thus they are not interoperable. To achieve semantic interoperability in INSPIRE data sets and services, authors propose the use of ontologies and Semantic Web technologies.

Data harmonization is important issue connected with the problem of achieving interoperability. It is a process of providing access to spatial data in a representation that allows for combining it with other harmonized data in a coherent way by using a common set of data product specification. Spatial data harmonization is performed by creating definition of reclassification and transformation rules (mapping schema) for source application schema. Mapping schema is then used for converting source data to target data, according to the output application schema. Creating those rules is very demanding and time consuming task which requires wide domain knowledge and detailed look into application schemas. Authors propose the use of application ontology created from application schemas and enriched by common domain and upper ontologies, which will integrate the domain knowledge and knowledge of data specification. By using methods of ontology mapping it is anticipated to support or even automate harmonization process.

Conducted project will take up issues associated with ontological representation of National Geodetic and Cartographic Resource and INSPIRE application schemas, data harmonization and integration. The main goal of maintaining semantic interoperability with use of Semantic Web technologies can be achieved with formal description of certain data models with the use of ontologies. Research will attempt to provide a new method of supporting or automating spatial data harmonization using ontological representation of selected application schemas, common upper and domain ontologies and various ontology matching methods.

Significance of the project

Proposed solution is a new approach for resolving problem of semantic interoperability between National Geodetic and Cartographic Resource and INSPIRE-based data sets and spatial web services. The use of commonly used Semantic Web technologies and ontologies guarantees compatibility with standards of representing meaning of published resources. Semantic annotation of spatial web services premises improvement in methods of service discovery. Presently, search process is based only on direct matching entered phrase to keywords and other properties which describe content of service. With the use of semantic metadata, a new method of “intelligent” search is available. Another opportunity which emerges from semantically annotated spatial web services is the possibility of service composition and orchestration. Process of service assemblage into chains can be automated with the use of “intelligent” mechanism. Semantic metadata can be used to select suitable geospatial web service for composition. Because of the used ontological approach and technologies, one of the expected results of the project is opening European infrastructure for spatial information to the world of Semantic Web.

In Poland, the problem of semantic interoperability connected with geospatial information was noticed but no further research was made to overcome this problem. Research of this issue proposed by this project will be a novelty in Poland.

Work plan

  1. Analysis of the application schemas of National Geodetic and Cartographic Resource and INSPIRE data sets.
  2. Research on available ontological modeling methods.
  3. Analysis of application schema to application ontology conversion methods.
  4. Review of available upper and domain ontologies for application ontology enrichment.
  5. Research on application ontology enrichment methods for creating semantically annotated application schemas.
  6. Evaluation of ontology mapping approaches and concept semantic similarity measurements for automated mapping between application schemas entites.
  7. Elaboration of automated spatial data harmonization and integration strategies with the use of Semantic Web technologies.