Catálogo de publicaciones - libros

Nowadays new applications ask for enriching the semantics associated to geographical information in order to support a wide variety of tasks including data integration, interoperability, knowledge reuse, knowledge acquisition, knowledge management, spatial reasoning and many others. Examples of such semantic issues are temporal and spatio-temporal data management, 3D manipulation, spatial granularity and multiple resolutions, multiple representations (providing different perspectives of the same information), vague and ambiguous geographic concepts, the relationship between geographic and physical concepts, and identity of geographic objects through time.

At the same time the recent years brought many developments that radically changed how we understand information processing. Data warehouses and OLAP systems have evolved as a fundamental approach for developing advanced decision support systems. This lead to improved data mining techniques allowing to extract semantics from raw data. Further, the success of Internet has generated a paradigm shift in distributed information processing leading to the area of Semantic Web, in which semantics is the fundamental component for achieving communication both for humans and applications. At the same time, mobile and wireless computing have entered everyone.s life through dedicated devices leading to location-based services. Finally, Grid computing, a paradigm enabling applications to integrate computational and information resources managed by diverse organizations in widespread locations, pushes the frontier of global interoperability. The fact that all these recent developments are entering the geographic domain increases the importance of the elicitation of the semantics of geographical information.

Geospatial conceptual data models represent semantic information about the real world that will be implemented in a spatial database. When linked to a repository, they offer a rich basis for formal ontologies. Several spatial extensions [5, 15, 17] have been proposed to data models and repositories in order to enrich the semantics of spatial objects, typically by specifying the geometry of objects in the schema and sometimes by adding geometric details in the repository. Considering the success of such 2D spatial extensions as well as the increased demand for 3D objects management, we defined a 3D spatial extension based on the concept of PVL already used in Perceptory and elsewhere. This paper presents 3D concepts and 3D PVL to help defining the geometry of 3D objects in conceptual data models and repositories. Their originality stems from the fact that no similar solution exists yet for real-life projects. The enrichment of the meaning of 3D objects geometries is discussed as well as its impact on costs, delays and acquisition specifications.

This paper proposes a method for evaluating the semantic similarity of GML elements (concepts). Due to the relevance of the relationship in the geographic context, it focuses on GML elements organized according to () hierarchies. It also proposes the method’s application to hierarchies, due to the semantics of the meronymic relationship within the geographic context, referred to as “place-area”. This semantics essentially concerns parts which are similar to and inseparable from the whole. A further contribution refers to the modeling of the hierarchy in GML. In particular, from the perspective of applying the approach to hierarchies, this paper proposes a method to represent and distinguish the concepts involved in the meronymic relationship within the element hierarchy.

In 2003 Pires [9] published a study that compared the results from the study performed by David Mark and Barry Smith with a similar study applied to Portuguese subjects. The paper concluded that the methodology of Mark and Smith, to establish ontologies from surveys of how users apply terminology, is applicable to identify conceptualization differences in GIS applications.

This paper is an extension of that work, presenting the results of the study in terms of university background. In response to series of differently phrased elicitations, 533 subjects (university students from several parts of Portugal and several academic disciplines) were asked to give examples of geographical categories. By this we statistically counted the most mentioned terms and related these to the university background.

The results were analysed in order to test the hypothesis: Students from different backgrounds have different conceptualizations of geographical categories due to their scholarly background. Our analysis refutes this hypothesis: students present the same examples for the presented categories and their disciplinary backgrounds cannot be shown to have an influence on the category choices.

Ontology has been conceived as a branch of metaphysics that studies the theory of objects and their relationships [3].

In this paper we aim to explore the relation Ontology/Geographical Information Systems from the cognition perspective. We raise the question; does scholarly background influence geographic categorization? In order to answer this question we used a survey that studied a specific set of geographic concepts, water bodies. The main reason behind the choice of these specific entities is that Portugal is a country exposed to the Atlantic and where water has been considered as an important element since the time of the Discoverers.

The survey is based on a similar approach taken in other parts of the world, such as England, Finland and others [11].

We propose to enhance a schema integration process with a validation phase employing logic-based data models. In our methodology, we validate the source schemas against the data model; the inter-schema mappings are validated against the semantics of the data model and the syntax of the correspondence language. In this paper, we focus on how to employ a reasoning engine to validate spatio-temporal schemas and describe where the reasoning engine is plugged into our integration methodology. The validation phase distinguishes our integration methodology from other approaches. We shift the emphasis on automation from the a priori discovery to the a posteriori checking of the inter-schema mappings. By doing so, we take advantage of the expressive power of the common data model in the source schema description and inter-schema mapping definition.

Conceptual models provide powerful constructs for representing the semantics of real-world application domains. However, much of this semantics may be lost when translating a conceptual schema into a logical or a physical schema. The reason for this semantic loss is the limited expressive power of logical and physical models. In this paper we present a methodology that allows to transform conceptual schemas while preserving their semantics. This is realized with the help of integrity constraints that are automatically generated at the logical and physical levels. As a result, the semantics of an application domain is kept in the database as opposed to keeping it in the applications accessing the database. We present such a methodology using the MADS conceptual spatio-temporal model.

Spatial configurations have a meaning to humans. For example, if I am standing on a square in front of a building, and this building has a door, then this means to me that this door leads into the building. This type of meaning can be nicely captured by image schemata, patterns in our mind that help us making sense of what we perceive. Spatial configurations can be structured taxonomically and mereologically by means of image schemata in a way that is believed to be close to human cognition. This paper focuses on a specific application domain, train stations, but also tries to generalise to other levels of scale and other types of spaces, showing benefits and limits.

Techniques and issues for the characterisation of an object-field representation that includes notions of semantics and uncertainty are detailed. The purpose of this model is to allow users to capture objects in field with internally variable levels of uncertainty, to visualize users’ conceptualizations of those geographic domains, and to share their understanding with others using embedded semantics. Concepts from collaborative environments inform the development of this semantic-driven model as well as the importance of presenting all collaborators’ analysis in a way that enables them to fully communicate their views and understandings about the object and the field within it. First, a conceptual background is provided which briefly addresses collaborative environments and the concepts behind an object-field representation. Second, implementation of that model within a database is discussed. Finally, a LandCover example is presented as a way of illustrating the applicability of the semantic model.

Collinearity is a basic arrangement of regions in the plane. We investigate the semantics of collinearity in various possible meanings for three regions and we combine these concepts to obtain definitions for four and more regions. The aim of the paper is to support the formalization of projective properties for modelling geographic information and qualitative spatial reasoning. Exploring the semantics of collinearity will enable us to shed light on elementary projective properties from which all the others can be inferred. Collinearity is also used to find a qualitative classification of the arrangement of many regions in the plane.

Gazetteer services are an important component in a wide variety of systems, including geographic search engines and question answering systems. Unfortunately, the footprints provided by gazetteers are often limited to a bounding box or even a centroid. Moreover, for a lot of non–political regions, detailed footprints are nonexistent since these regions tend to have gradual, rather than crisp, boundaries. In this paper we propose an automatic method to approximate the footprints of crisp, as well as imprecise, regions using statements on the web as a starting point. Due to the vague nature of some of these statements, the resulting footprints are represented as fuzzy sets.