1

I am currently interested in representing some geo-data in a semantic way via RDF/OWL to allow formal reasoning on a geometry level. However I am lacking a good overview on the current standards to do so and on the benefits and limitations of each.

If you could recommend a good overview webpage or paper I would be very thankful!

  • Presumably, you know about Open Geospatial Consortium (OGC) opengeospatial.org, Geography Markup Language (GML) en.wikipedia.org/wiki/Geography_Markup_Language, and Simple Features en.wikipedia.org/wiki/Simple_Features – Martin F Apr 5 '14 at 1:32
  • Sorry to be picky. You capitalize "GeoData" as though it is a specific proper noun. Do you mean "geo data" the general notion? Perhaps you could provide a little more detail on specific topics you are considering. (The first thing i thought of was my answer below but i may be completely off the mark.) Also, should "webpage oder paper" be "webpage or older paper"? – Martin F Apr 5 '14 at 1:40
1

This is rather old (1994) so predates RDF/OWL by a long way, but may be of theoretical interest. It was developed independently from, but around the same time as, Geographic Markup Language (GML), and is more of a conceptual discussion than a comprehensive catalog.

Towards a geographic semantic database model, by M Feuchtwanger (me). Here's the abstract:

The management of geographic data is a great problem in contemporary cartography. To date, little theory has been developed to assist such a task. This thesis proposes a geographic semantic database model: a concept for the design, construction, and use of geographic databases. The work involved the synthesis of both general semantic database concepts and specific geographic information concepts.

A logical database model incorporates notions of the structural and behavioral aspects of stored information. Structurally, a database contains entities, relations, and attributes. Behaviorally, a database has queries and transactions. Database models are evolving from syntactic to semantic forms, representing greater ability to directly and easily model reality.

Any things of interest in geographic data processing can be called phenomena. A phenomenon exhibits three primary characteristics: topical, spatial, and temporal. That is, it has some identification and position, and exists at some time. Information on phenomena thus exists within three characterization domains. It also exists within three abstraction domains: generalization, realization, and construction. That is, geographic data have some accuracy and resolution, some form between reality and concept, and a level of meaning or applicability. The characterization and abstraction domains are the particularly geographic ways for logically partitioning a collection of data.

The proposed geographic database model contains entities, such as features, profiles, layers, and composites, which represent geographic phenomena. The entities are characterized by topical, spatial, temporal, and scale attributes, and by semantic, topologic, and abstraction relations to other entities. They can be retrieved, displayed, or updated by database manipulations comprising selections and actions. The entities also exist at different levels of abstraction: at different scales, appropriate for different levels of investigation; in analytic or graphic form, depending on whether they are to be used for machine or visual processing; and as applied, basic, or primal constructs, appropriate for different levels of use.

Combining concepts from database management and analytical cartography into a geographic database model not only facilitates the analysis and design of geographic databases but also is a step towards a general theory of geographic information management and analysis.

It's available as a PDF scan or PostScript files.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.