Poster at GIScience 2006

This is the poster I presented at GIScience 2006. The poster forms part of a proposal to consider how the Object-Field model, a model that is considered to combine both the discrete-object and continuous-field views, has some unique qualities for collaborative decision making. It aims to prompt conceptual and theoretical thoughts and discussions by identifying when to use the Object-Field model and not the conventional object and field models . The insights and comments addressed the following kinds of questions :

1) Is the integration of knowledge and observational data useful & usable ?

2) Is the integration of Object & Field views brings new analytical advantages?

3) Is the Object-Field model an improved sharing of collaborative analysis means?
An answer?
The Object-Field model enables visualization and representation of objects in a field. This approach (visualization of field objects) could work better than the conventional approaches in a collaborative environment by presenting the users with multimedia objects, objects that record information in text, image or other forms, related to specific locations or a set of locations in a field. An object in this context is a modeller’s conceptualization/knowledge. Thus, the users are not presented with just observational data but this data is augmented by visualized users' conceptualizations/knowledge of geographic domains, and their understanding using embedded metadata expressed as semantic and uncertainty objects. These embedded semantics and uncertainties propose a new dimension in the metadata discussion. This is the explicit recording of collaborative understanding, interpretation and criticism.

In the Object-Field model, the four types of relationships between locations and objects enables the conventional one location-one object and many locations-one object relationship but it also enables the one locations-many objects and many locations-many objects. This mathematically-based relations between objects and locations enables the user to rearrange the objects in a way that supports comparisons. It also enables a object-based approach when operations are applied. Thus different conceptualizations and interpretations between collaborators can be easily visualized and cross-checked by reapplying them to new purposes or procedures

Integrating observational data and derived knowledge (expressed as metadata objects in the Object-Field model) enable us to improve sharing of analysis by designing a user interface that uses the same set of tools for the exploration of data and knowledge. This integration is important in cross-cultural collaborative environments as it manages semantic inaccuracies and making metadata not only useful but also usable.

Definitions of Object-Field, Field and Object

The Object-Field view of the world is considered to combine of both the discrete-object and continuous-field representations. The Object-Field model attempts to integrate the field and the object view in a single, combined and integrated data model. This is achieved by mapping locations in a field to objects. Aggregating field locations forms the objects. ). This model uses a single elementary spatial unit (hereafter object element) to exploit the benefits of continuous-field and discrete-object views. The object elements are associated with a field value and a variable number of object references. See an Object-Field example by Cova & Goodchild (checked 14 October 2006).

the discrete-object view of the world is considered as a series of entities located in space. An object is a digital representation of these entities. Objects are classified into different object types such as point objects (stores), line objects(retail network) and area objects (London Boroughs). These Objects are defined by their boundaries. In turn, we attach/associate one or more attributes with these objects to specify what is located at these places. These general classes are instantiated by specific objects and, we can attach behaviours to these objects

The continuous-field view of the world is made up of properties varying continuously across space. The key factors of the field view are spatial continuity and self-definition. As the key characteristics of the field view is spatial continuity and self-definition we are not forced to identify objects and their boundaries. In other words, the field is a collection of a certain kind of measurements (such as consumer’s spending behaviour) that are used to define a value everywhere in the field and it is the values themselves that define that field.

My publications about the Object-Field Model, collaborative Visualization and Decision Making

Voudouris, V., P. F. Fisher and J. Wood (2006) 'When and Why Object-Fields and not just Objects or just Fields?'. Presented at the Fourth International Conference on GI Science 2006 (Germany), Ifgi-Prints Series.

Voudouris, V., P.F. Fisher and J. Wood (2006) 'Capturing Conceptualization Uncertainty Interactively using Object-Fields' in W. Kainz, A. Reid and G. Elmes (eds) 12th International Symposium on Spatial Data Handling. Springer-Verlag.

Voudouris, V. P.F. Fisher and J. Wood (2006) 'Collaborative Visualization: Metadata within Object-Fields as Communication Means'. Presented at the RGS-IBG Annual International Conference 2006, London, UK.

Voudouris, V. and S. Marsh (2006) 'Geovisualization and GIS: A Human Centred Approach'. In Visual Languages for Interactive Computing: Definitions and Formalizations (Eds, F.Ferri), Idea Group Inc.

Voudouris, V., J. Wood and P.F. Fisher (2005) 'Collaborative geoVisualization: Object-Field Representations with Semantic and Uncertainty Information' in: Meersman, R., Tari, Z., Herrero, P., et al (Eds) On the Move to Meaningful Internet Systems OTM 2005, Lecture Notes in Computer Science (LNCS), Vol 3762, Springer, Berlin

Welcome

The purpose of this blog is to promote discussions about the Object-Field model, applied statistics and mathematics, decision making, visualization, object-oriented modeling and knowledge representation.

From time to time, I will post my personal opinion about these issues based on my research and work experience.

Please post interesting ideas, links and articles about the data and knowledge modelling, applied statistics and mathematics, theory of decision making, visualization and object-oriented modeling.