Sunday, 3 February 2013

Never Too Late: CyberGIS, Semantic Web, Event Ontology

Will the event ontology link CyberGIS and Semantic Web in a new collaborative vision? After working on geospatial and semantic web issues for years, I can not help but wonder this question. The CyberGIS now exhibits a new dynamism: that is an ever more complex framework between cyberinfrastructure, GIS and spatial analysis, which comprises at least three visions (Wang 2010; Wang et al 2012):
click here for full slide
  • a cyberinfrastructure of user‐centered and a knowledge‐sharing vision;
  • a collaborated geospatial computing by integrating location‐and‐place information;
  • the computational intensity of spatial analysis for solving problems, especially in facilitating partnership of different domains within various distributed organizations.

What are exiting relations between CyberGIS and Semantic Web?

First, Semantic Web technologies provide large datasets with structured machine‐processable formats that can convey human meanings as a semantic layer for cyberinfrastructure. This in turn can be observed by a recent trend of spatial cyberinfrastructure movement to the Linking Open Data (LOD). Efforts include developing ontologies, publishing and connecting data via URIs, or provide reasoning services on the Web. Central to this issue is that W3C projects has been participated by governments's efforts in releasing datasets in RDF formats and linking them to other datasets for citizens use. Open government datasets are integrated with semantics and structure, and as a result, this movement provides CyberGIS with additional resources as well as user contexts for knowledge‐sharing. (Janowicz et al 2010; Shadbolt et al 2012; Hendler et al 2012; W3C 2012)

Additionally, with the increasing development of semantic web tools for user participation, citizens now can participate in contributing and correcting data, thus the complementary role of Volunteered Geographic Information (VGI) to CyberGIS has more potential within the framework of Semantic Web. (Goodchild 2007; Poore 2011; Elwood et al 2012)

Second, geospatial dimension of CyberGIS may provide linking datasets an intuitive concept for government angency to classify, display and curate data. For instance, in UK, Ordnance Survey has made efforts in connecting Linked Data and releasing Regions and Named Places through OpenUpLabs Datasets. In US, the makes linked data available to general public, and has integrated its government geospatial services, Geospatial One Stop, with semantic technologies. Other governmental infrastructure cases, to name just a few, include Scotland statistical geography, FAO geopolitical ontology of UN systems, Geological Survey of Austria (GBA)‐Thesaurus, National Geographic Information Institute of Korea, and the GeoLinkedData of the National Geographic Institute of Spain. (Shadbolt et al 2012; Hendler et al 2012; W3C 2012)

Above linking spatial data infrastructure by governments’ movements are not alone. Online communities like OpenStreetMap also adds a spatial dimension to the Web of Data (i.e. LinkedGeoData); Yahoo Geoplanet RDF dumps Yahoo! GeoPlanet datasets in order to bridge the gap between the real and virtual worlds. GeoNames relates geographical names to geospatial features through GeoNames ontology. (W3C 2012)

Geospatial applications using event ontology and modeling via cause relation

(1) Event ontology in geospatial applications

Ontological efforts have been made in GIScience for years. Schematic representation of different approaches and methods such as top‐level and domain specific ontologies are generally reviewed in the survey of Agarwal (2005). Among the top level concern for human semantics, event concept is straightforward. In the field of classic geography, most research focus on modeling objects or fields incorporated with temporal dimension.
Grenon and Smith (2004) propose SNAP and SPAN as basic formal ontology which is incorporated event concept with dynamic spatial ontology. Furthermore, Worboys (2005) has argued that event‐oriented approaches should be upgraded to an equal status with ‘things’ in representations of dynamic geographic phenomena. For the Geospatial Event Model (GEM) developed by Worboys and Hornsby (2004) has set event as the major concept, and each geospatial object or event in an information system is situated in a setting in which the spatial, temporal, or a combination of both concepts can be modeled.

Nevertheless, even the context of CyberGIS is slightly different in its scale from general GIScience, it is similar in domain applications such as disaster management. Nowadays event ontology has been employed and demonstrated in several practical cases. For instance, GeochangeEvents (, a knowledge base linked within Linking Open Data Cloud, describes events of earthquakes, volcanoes and tsunamis collected from NOAA. The USGS has also made efforts in building ontology for The National Map (version 3.0) to fulfill the purpose of linked‐data cyberinfrastructure. In particular, they use the feature and event based topographic vocabularies for specifying triple data semantics (Usery and Varanka, 2011).

(2) Event‐based approach with cause relation for modeling method

Early research on cause relation in modeling event focused on temporal relation of event logs. For example, "precede" and "follow" are casually related in state changes when modeling composite events (Chakravarthy, 1994). Similarly, EventNet uses cause and effect to infer temporal relations between commonsense events, therefore it makes predictions of preceding or subsequent events possible (Espinosa and Lieberman, 2005). Lexicalization of concepts for human language like WordNet also considers casual relation like “show and see” or “raise and rise”. The verb here is mainly to encode event or state within temporal contexts (Fellbaum 2010).

Whereas conventional research explained casual relation primarily for temporal purposes, ontological event-based research looks to more abstract event concept for modeling methods. For instances, the design patterns in Event Model F propose the causal pattern and effect pattern to relate two events that have a common cause (Scherp,2009) . Or specific to the spatial perspective, event has a unique position like spatial position of objects, and event can be identified because the framework of causal relations (e.g., in Worboys’s example that if a car saw a yellow light the car will brake and slow down. This is one event that occupies the same spatiotemporal region. However, if from the perspective of the cause relation, this event will be identified as two events: the braking event as the cause and the slowing event as the effect) (Worboys, 2005).

Although the modeling method of cause relation has been challenged in literature, equally important is the assumption that the cause relation provides possible solution to decompose event into atomic elements for machine reasoning. The work related to ours mainly fits into the use of the "action" in Geospatial Event Model, in which “action” is subcategory of event; and at least one object or one agent acts to produce a particular event. in particular, Worboys (2005) has made this point more clearly on a comparison analysis of event, action and process by situation calculus and interval temporal logic.

After relating CyberGIS, semantic web and event ontology, there is a general buzz of excitement in these potential relationships. In sum, cyber-infrastructure people may be more cautious on whether or not to dive on in semantic web because infrastructure based tasks are generally relating to big data issues, and this in turn, not only involves technical issues but also big financial budgets, social and political concerns as indicated in the recent CODATA 2012 conference. But whatever the barriers, we should definitely take advantage of the potential possibility of these three. One of our arguments for relating the critical information of disaster ,for decision makers and general public, is to harness the event concept that can help our data computation and human thinking in a more collaborative and productive way.


  1. Agarwal, Pragya, Ontological considerations in GIScience, International Journal of Geographical Information Science, Vol.19, no. 5, 501-536, 2005
  2. Chakravarthy, Sharma and Krishnaprasad, V. and Anwar, Eman and Kim, S.-K., Composite events for active databases: Semantics, contexts and detection, Proceedings of the Twentieth International Conference on Very Large Data Bases, 606-617, 1994
  3. Elwood, Sarah and Goodchild, Michael F. and Sui, Daniel Z., Researching Volunteered Geographic Information: Spatial Data, Geographic Research, and New Social Practice, Annals of the Association of American Geographers, Vol. 102, no. 3, 571-590, 2012
  4. Espinosa, Jose and Lieberman, Henry, EventNet: Inferring Temporal Relations Between Commonsense Events, MICAI 2005: Advances in Artificial Intelligence, Lecture Notes in Computer Science, Vol. 3789/2005, 61-69, 2005
  5. Fellbaum, Christiane, WordNet, Theory and Applications of Ontology: Computer Applications, Chp.10, 231-243, 2010
  6. Goodchild, Michael F., Citizens as Voluntary sensors: spatial data infrastructure in the world of Web 2.0, International Journal of Spatial Data Infrastructures Research, Vol. 2, 24-32, 2007
  7. Grenon, Pierre and Smith, Barry, SNAP and SPAN: Towards Dynamic Spatial Ontology, Spatial Cognition and Computation, Vol. 4, no. 1, 69-104,2004
  8. Hendler, James and Holm, Jeanne and Musialek, Chris and Thomas, George, US Government Linked Open Data:, IEEE Intelligent Systems, Vol. 27, no. 3, 25-31, 2012
  9. Krzysztof Janowicz, Sven Schade, Arne Bröring, Carsten Keßler, Patrick Maué, Christoph Stasch, Semantic Enablement for Spatial Data Infrastructures, Transactions in GIS, Vol. 14, no. 2, 111-129, 2010
  10. Poore, Barbara S., Users as essential contributors to spatial cyberinfrastructures, Proceedings of the National Academy of Sciences of the United States of America, Vol. 108, no. 14, 5510-5515, 2011
  11. Scherp, Ansgar and Franz, Thomas and Saathoff, Carsten and Staab, Steffen, F—A Model of Events based on the Foundational Ontology DOLCE+DnS Ultralite, Proceedings of the Fifth International Conference on Knowledge Capture, 137-144, 2009
  12. Shadbolt, Nigel and O'Hara, Kieron and Berners-Lee, Tim and Gibbins, Nicholas and Glaser, Hugh and Hall, Wendy and Schraefel, M. C, Linked Open Government Data: Lessons from, IEEE Intelligent Systems, Vol. 27, no. 3, 16-24, 2012
  13. Usery, E. Lynn and Varanka, Dalia, Design and development of linked data from The National Map, Semantic Web, 1-14, 2011
  14. W3C, Linked Data Sets (i.e., with Dereferenceable URIs) available as RDF Dumps, 2012
  15. Wang, Shaowen, A CyberGIS Framework for the Synthesis of Cyberinfrastructure, GIS, and Spatial Analysis, Annals of the Association of American Geographers, Vol. 100, no. 3, 535-557, 2010
  16. Wang, Shaowen and Wilkins-Diehr, Nancy R. and Nyerges, Timothy L., CyberGIS - Toward synergistic advancement of cyberinfrastructure and GIScience: A workshop summary, Journal of Spatial Information Science, no. 4, 125-148, 2012
  17. Worboys, Mike, Event-oriented approaches to geographic phenomena, International Journal of Geographical Information Science, Vol. 19, no.1, 1-28, 2005
  18. Worboys, Michael and Hornsby, Kathleen, From Objects to Events: GEM, the Geospatial Event Model, Geographic Information Science, Lecture Notes in Computer Science, Vol. 3234/2004, 327-343, 2004
Citation Information: 
Andrea Wei-Ching Huang (2013), Never Too Late: CyberGIS, Semantic Web, Event Ontology, URL:

Creative Commons LicenseNever Too Late: CyberGIS, Semantic Web, Event Ontology by Andrea Wei-Ching Huang is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.