Kristin Stock
Research Fellow in Geospatial Semantics, Faculty of Social Sciences
Contact
Biography
Dr Kristin Stock originally trained as a surveyor, completing a Bachelor of Science at the University of Otago (New Zealand) in 1989. She then worked as a surveyor in a nickel mine in North Queensland and in a private surveying practice in Bundaberg, Queensland. During this time, she also completed a Graduate Diploma in Urban and Regional Planning, a Graduate Certificate in Computing and a Graduate Diploma in Surveying Practice.
In 2000, she completed a PhD in Geographic Information Systems, focussing on semantic integration and translation of heterogeneous spatial databases. She also worked as a Senior Research Assistant on issues of cadastral representation in GIS, and as a Systems Architect with the Australian Capital Territory Planning and Land Authority. In that role, she designed a land information system for the territory for the management of a wide range of cadastral, land administration, planning, surveying and feature names data, lead various projects involving design and development activities, and was involved in international standardisation efforts. She then worked as a senior consultant for Social Change Online, and now runs her own consultancy (Allworlds Geothinking) while working part-time at the Centre for Geospatial Science.
Expertise Summary
My work involves the development of methods to allow people to share geographic information across the globe. Such information is often highly variable and difficult for users to interpret, and the methods I am developing involve representing the meaning of geographic information so that it can be understood by remote users. These representations can then be used by humans and computers so that they can analyse and manage the information intelligently. The representations I am working on incorporate techniques in geospatial semantics and ontologies research, and extend this work to allow semantics to be used for automatic analysis and data mining.
Research Summary
Kristin's research within the field of geospatial semantics pursues a number of different directions:
1. Natural Language and Multi-Lingual Geospatial Semantics
This work involves the development of methods to allow users to intuitively interact with geographic information through natural language, and also aims to support a multi-lingual environment. The goal is to allow users to retain their own unique, dynamic and context sensitive ways of conceptualising the world, and to interact with geographic information systems (GIS) on that basis, rather than being required to speak the language that is most convenient for the GIS.
Kristin's recent work in this area focuses on the use of linguistic semantic primitives from Natural Semantic Metalanguage (NSM) to express the semantics of spatial objects and spatial relations. She is working on a formal grammar of NSM to support validation of user input, and is developing methods to allow users to describe spatial relations (e.g. CONTAINS, ON, IN, CROSSES) using their own conceptualisations in a particular context, and to automatically interpret those descriptions and perform semantic matching. Kristin has conducted various qualitative experiments with participants to support this work, and incorporates research from linguistics, cognitive science, geography and information systems. Some of this work is being undertaken as part of the EU funded EuroGEOSS project.
2. Deep Geospatial Semantics
This work is in its early stages and is motivated by the superficial nature of current representations of semantics. Such representations include ontologies, and are characterised by a model of semantics based on categorisation and consensus viewpoints. While these approaches have their benefits, they do not capture the deeper ways in which people think about the world and the environment with which they interact. This research involves the development of representations of deep geospatial semantics using logics, including modal logics. This work also makes use of NSM techniques, and incorporates qualitative research methods for extraction of information about human worldviews and attitudes.
3. Knowledge Infrastructures
The research involves the development of knowledge infrastructures to support the use of geographic information by scientists in the scientific process, and involves various collaborations with Edinburgh, Cardiff and Muenster Universities, with particular focus on the evolution of knowledge and the discovery and visualisation/representation of patterns in the relationships between knowledge artefacts. Recent activity on this project includes the Compass Project.
4. Semantic Registries for Spatial Data Infrastructures
This research involves the design of registries or catalogues for spatial data infrastructures that include semantic information. Work already completed represented these semantics using Feature Type Catalogues, and included OWL ontologies in registries to assist in intelligent querying, service composition and automated interpretation of web services and data resources.