Man seeks to understand the essential universalities of the universe. Uncovering a universal ontology of geographic space contributes a further piece to this universal mosaic.
The idea of a universal (top, standard) ontology of geographic space (geospatial ontology, geo-ontology) as a universal semantic reference system presents a major scientific challenge that prompts several questions. Our goal here is to discuss the geographic space, limited to the planet Earth, and the possible applications in a number of other spatial-domain tasks. The aim of this book is to answer the following questions: “is it reasonable to design a universal ontology of geographic space and to expect it to be applied?” and “how could such an ontology be implemented in professional and general society through distributed information systems?” We will assume that any semantically weak spatial data can be enriched to gain a more efficient interoperability. The universal ontology of geographic space as an enrichment engine of spatial data semantics is conceived as an interdisciplinary task. While there is a vast amount of literature on the semantics of spatial symbolism and the various problems of developing ontologies of geographic space, the correlation in database models, the realities of space, the concepts in the human mind, and the meaning of symbols/words are virtually hidden. For example, data management and analyses are often limited to the concepts of hardware, software, and standards. This book attempts to challenge the computer-driven dominance over the topics of semantics in the “geo-” domain that also covers the pillars of philosophy, logic, and linguistics as intrinsic disciplines requiring an understanding of spatial issues. This publication integrates the complexity of spatial dimensions in geographic space with the philosopher’s concern for ontology, and furthermore the universal ontology of geographic space.
One of the aims is to escalate the current scope of research to support the development of semantically interoperable systems of geographic space. Human activities are based on extracting meaningful information within optimised and advanced applications, to form widely complex, obscure, or fundamental knowledge. These outputs may be enriched by incorporating philosophy, via (technological) implementations, as a consideration in order to reach more effective computer applications that are often beyond traditional approaches. For example, the problem of ontology covers many interesting topics, e.g., semantic enrichment, the development of a semantic reference system, semantic similarity measures, data integration, interoperability, usability, knowledge management, applications of semantic web, interface design for spatial semantic content. This publication, however, attempts to unfold the complex relationship between the spatial dimensions and the philosophical concepts.
A universal approach to the ontology of geographic space has already been and is going to be a comprehensive task for establishing more effective spatial models. The concept of a universal spatial ontology should be independent of location, culture, and time. It should be fundamental and universal in the same way that the number p defines the ratio between the diameter and the circumference of a circle. The term “universal” therefore means all embracing and for general propose. The process of generating a universal ontology is subject to generalisation processes as well, compared to models or maps. To make a model that depicts the entire reality is too complex and as such almost incomprehensible in natural language. Practically, a universal conceptual model should comprise just the generalised, essential spatial characteristics of reality, semantically enriched with patterns and the vocabulary of databases, depending on the context. The most known generalisations in the spatial domain are the cartographic and geometric rules.
Many upper-level ontologies describe quite general concepts which are the same across all knowledge domains and cultures. They demonstrate different approaches to their reference ontology, but also many similarities and universalities—as rules are valid across them. They can be harmonised to a certain common level by improving the proactive communication between different scientific disciplines. Sowa suggests that, except for perhaps Aristotle’s original ten categories, there has never been any proposal for a universal ontology that has been widely accepted by anybody other than the author and a limited number of colleagues. Furthermore, there are strong reasons for doubting that a universal ontology can ever be achieved—at least not until all major issues in science have been definitively resolved. Since our conscious experiences seem to be individually limited, and for common understanding they should be holistic, multi-level, and global, the universal ontology should therefore be relevant to general and specialised goals. However, the human being has the ability to learn (almost) anything new, which is not something that current artificial intelligence can do, therefore the creation of a universal ontology would be highly appreciated.
The universal ontology of geographic space as an enrichment engine of spatial semantics of geoinformatics is reflected in the interdisciplinary style of this book. We assume that our readers come from the “spatially” nonspecific backgrounds, such as informatics, ontology, semantics, interoperability, quality of data, theoretical linguistics, cognitive science, etc. Nevertheless, the most obvious target-reader would be from one of the spatial science fields, in particular, geoinformatics. In addition, this publication will be of interest to university lecturers and professors, students, researchers, and developers of the various kinds of spatial applications.
The book consists of ten chapters and is organised into two sections. Section 1 is entitled “Towards a Universal Ontology in the Spatial Domain,” and comprises five chapters that consider the concepts of reference ontologies and aim towards a universal ontology of geographic space. Section 2, entitled “Applied Ontologies in the Spatial Domain,” comprises a further five chapters, in which the concepts of data enrichment are discussed. Section 1 is oriented toward fundamental approaches, while Section 2 comprises various applications which are correspondingly supplemental to the theoretical part, as follows in the detailed description.
The first chapter is entitled “A Universal Geospatial Ontology for the Semantic Interoperability of Data: What are the Risks and How to Approach Them?” The term “Ontology” has been borrowed from philosophy, and refers to the science of describing the kinds of entities in the world, and how they are interrelated. The conception of ontology has been defined in a wider context where the ontologies are considered to be the backbone technology for communication between agents. Here the term “ontological commitments”, i.e. agreements between a set of agents, uses a shared vocabulary in order to communicate about the domain of discourse (e.g., classes, relations, functions, or other objects). Ontology represents the concepts according to the various levels of abstraction. It can be classified into three types: (1) an upper-level (top-level, global) ontology describes general concepts which are independent of a specific domain or application, i.e. those that are essential for a human understanding of the world (e.g., space, time, and action); (2) a domain ontology provides the vocabulary of the concepts and the relationships that depict the theories of a given domain (e.g., an ontology of the concepts relating to the medical field); (3) an application ontology describes the concepts relating to a particular application (e.g. the ontology of the national topographic database describes concepts relating to topographic phenomena). A universal ontology describes reality at a higher level of generality. It is a more general-purpose ontology that could be generated from several domain-specific ontologies. A universal ontology can be defined as a common, foundational, higher-level ontology that describes reality, its nature, composition, structure, meanings, and classes. Generalisations are usually based on domain ontology. However, such an ontology could be insufficient to generalise all elements of heterogeneous data models. Ideally, generalisations should be based on a universal common description of the data and should entail a greater commonness amongst contexts. The idea of a universal ontology of geographic space has triggered many research efforts to investigate the use of such an ontology, i.e. to resolve the problems of semantic heterogeneity and the uncertainty in meaning of geospatial data. However, the universal ontology of geographic space refers to a common upper-level ontology that describes real world phenomena based on widely consensual agreements in the GIS community.
“Geographic Space Ontology, Locus-Object, and Spatial Data Representation Semantic Theory.” The second chapter explores the ability of morphogenetic modelling in the recognition of a major ontological and semantic concept of geography, namely, the locus-object. The locus-object couple concept results from the interrelation formalisation between the geographical locus, the geographical object and the geo-localisation notions. Geographic loci and objects are linked and both geo-localised. The links and relations between locus and object are mathematically formalised by geospatiology in the study of the logical role of space in the study of entities on the surface of the Earth. Morphogenetic modelling recognises the loci of the geographic space. By convention, the sets of loci and the objects form a Cartesian product, i.e. the elements of these sets are distinct ordinate couples, each pair consisting of a locus and an object. If the locus and the object are both differentiated, the couple is strongly differentiated by the locus and the object. If the object is differentiated, but not the locus, then the couple is weakly differentiated by the object. And, if both the locus and object are undifferentiated, then the couple is undifferentiated. Abstractly, in economic theory, a homogeneous “plain of transportation”, where you can move at the same cost equally in all directions by generating transport networks of identical forms is a geometric space that is not differentiated by their loci or by objects (see the summary of chapter 10 below). Identifying universal ontologies of geographic space supposes, firstly, the ability to recognise, extract and visualise loci, objects and localisations in geomatics systems.
“Toward an Architecture for Enhancing Semantic Interoperability Based on the Enrichment of Geospatial Data Semantics.” A universal ontology of geographic space would play an invaluable role, given that reference ontologies in GIScience are considered standard, which supports the translation between different GIS or the negotiation of meanings between communities that use different vocabularies. Despite the fact, that upper-level ontologies aim to be “universal”, the several upper-level ontologies of the geographic space that were developed nevertheless differ notably with respect to scope and categorisation. The existing upper-level ontologies for the geographic space demonstrate that the development of a “universal” ontology of geographic space is exceptionally hard to achieve due to the complexities of geographical categories and the lack of consensual agreement on the basic set of geographical concepts. This chapter argues that semantically weak geospatial data can be enriched to enhance semantic interoperability. It proposes a conceptual architecture designed to support enhanced semantic interoperability in dynamic networks that focus on semantic enrichment. It includes a coalition management module, an ontology enrichment module, and a semantic mapping module. The modules perform different types of semantic enrichment and can be employed to support various semantic interoperability tasks. Within the different enrichment methods, the role of upper-level ontologies is explained; it is argued that they play a key role in the semantic interoperability framework. Network analysis is proposed as a form of semantic enrichment, since it allows for the discovery of new semantic structures (coalitions) which are not a priori perceptible, and which convey a meaning for users. The principle behind the suggested method is the computation of a semantic proximity measure between node contexts. This proximity is called semantic attraction, because it indicates how two nodes attracted to each other become part of the same coalition in virtue of common or similar context parameters. Eventually, when coalitions are formed, the coalition contexts are computed by merging the contexts of the members of the coalition.
“Geographical Process Representation: Issues and Challenges.” The fourth chapter addresses the issues and challenges that arise when building a general spatio-temporal ontology for representing and reasoning geographical processes as a part of a desired universal semantic reference system of geographic space. It examines the foundations and formalisms upon which the development of such an ontological model of geographical processes can be based. The development of a universal semantic reference system of geographic space is an intricate research challenge which raises numerous issues. This chapter surveys the object- and field-based views of spatio-temporal data representation and provides an insight into the challenges of developing a comprehensive ontology of dynamic geographic phenomena that is able to manipulate both views. Objects are defined by some abstract notion of identity which (1) define its spatial extent at any one time and (2) enable it to persist through changes in spatial location and other attributes. Fields are generally represented by decomposing geographic space in minimal regions and then assigning to each region a value from a certain range. A characteristic of geographical information is that it may be affected by vagueness, leading to additional representational difficulties. Vagueness in geographical processes is intimately related to the granularity of spatial and temporal information, and different interpretations may arise depending upon the level of granularity at which a process is observed. Spatial granularity has been treated in geographical information science and in the field of spatial data mining to improve the capability of the system to work appropriately with different map scales and to develop clustering algorithms which group spatial regions according to a set of characteristics (spatial or not). An aggregation or agglomeration consists of bringing together a group of individual objects so that they can be considered as a single object.
“Human Cognition: People in the World and World in Their Minds.” This chapter focuses on the evidence of the existence of a universal ontology of geographic space, based on the study of a psychological approach to the human cognition and spatial representations from different cultural and temporal backgrounds. Designing the universal ontology requires a detailed understanding of the fundamentals of the real environment and its processes. It involves not only entities and their actions, but also the processes of human perception. Our intelligence makes possible the creation of mental models of objects or events which simulate and predict the possible successes of our actions. The global ability to map read is composed of particular abilities, e.g. shape discrimination, decoding symbols, mental rotation, as well as the ability to orientate in geographic space, which is itself composes product of other cognitive abilities. From the point of view of developmental psychology, it is necessary to respect the developmental stages of the user’s group during the creation of a universal ontology of geographic space. Despite the fact that, across different societies, people use various means of spatial thinking and understanding, there exist similarities that help to overcome these differences and to understand the intended meaning. This fact supports the existence of a universal ontology. It is important to note that a universal ontology of geographic space must be based on these similarities, and that identifying these cultural cross-points will be one of the fundamental tasks in the process of designing a future universal ontology.
“Representing Geospatial Concepts: Activities or Entities?” A knowledge representation of geospatial entities is dependent on an ability to share their structural properties along with their functional properties which define their usage for human society. Ontologies of geographic space have mainly relied on taxonomy-based and mereology-based ontologies. While the structural properties of entities, such as shape, topography, orientation, or mereology, are considered to be important tools for an ontology of geographic space specification, the existence of a structural property is not a sufficient condition for the existence of any function. Contrastingly, a parallel approach assumes the independent existence of function- or action-based concept hierarchies, and builds on the premise that human-activities associated with any given geospatial entity are essential for the specification of the entity concept itself. The challenge of integrating the two perspectives is increased by the constraints of ontology specification techniques and is also inhibited by the difficulties in specifying probabilistic knowledge in ontologies and heuristics. These constraints limit the possibility of specifying partial knowledge of the linkages between geospatial entities and geospatial actions. Geospatial concepts are no exception and ontologies of geographic space have been employed to specify concepts in the geospatial domain. The knowledge representation of concepts rests on a similar foundation, provided by the theory of semantics in human communication. Three major utilities of ontologies include: (1) communication between people; (2) interoperability between computer systems; and (3) system-engineering related. The important question that needs to be addressed in the area of ontologies of geospatial entities is how one can verify ontologies. A combined approach would require two types of verification: the categorisation of all known instance and checking for infeasibility and the evaluation of truths.
“The Ontology Engineering Method for the Integration of Building Models: The Case of IFC and CityGML.” Three main uses of ontologies may be observed: (1) assisting communication between human beings; (2) achieving interoperability between software systems; and (3) improving the design and quality of software systems. Geospatial applications are often based on spatial objects as well as spatial relationships. The information can be represented as a geometric model that defines the geometric objects and elements types, and as a semantic model that defines the entities and their non-spatial characteristics and the relationships among the entities. This chapter focuses on the integration of the two most prominent semantic models for the representation of Building Information Modelling (BIM) and geospatial objects, namely, Industry Foundation Classes (IFC) and City Geography Markup Language (CityGML). Professionals from both domains have made significant attempts to integrate these models in order to produce useful common applications. Previous experience suggests that the best approach for achieving interoperability between the IFC and CityGML models is to integrate geometric models and to harmonise semantics. This chapter focuses on the semantic integration of the IFC and CityGML building models for bidirectional conversion. The IFC and CityGML use different terminologies to describe the same domain and there is a great heterogeneity in their semantics. In reality, there are major differences in systems, formats, applications, data schemas and the quality of the spatial data that is produced by or used by different stakeholders. Through the reference ontology, the proposed method contributes towards a formal mapping between the IFC and the CityGML ontologies that allows for a bidirectional conversion between them.
Furthermore, the method provides an approach in the direction of the complete integration of the CityGML and the IFC through the development of a Unified Building Model (UBM) that considers different classes, attributes and relations.
“Semantically Enriched POI as an Ontological Foundation for Web-Based and Mobile Spatial Applications.” Ontologies describe only parts of the universe. They represent a limited world view. There is no difference between the term “ontology,” and the term “mini world,” used in database theory. Mini worlds represent only small parts of the complex, universal world. Instead of modelling the entire natural and human entities, their relationships, as well as interferences between humans, nature, and laws, it is advisable and pragmatic to select only parts of it and describe these in a manner that fits the requirements of the user. A universal ontology would serve as a foundation of each existing concept of geographic space and therefore—in a formal version—as a foundation for nearly all computerised spatial applications. An idea like this represents a top-down-approach: the universal ontology stands at the highest level of abstraction, whereas all concepts and models in whatever spatial field are derivations. While developing a web-based travel planning system, the toned for a mobile component was identified. Such a system enables users to plan a trip in advance using a web-based application, and also to modify the original plan wherever and whenever they want while making the journey. Within both components, Point-of-Interest (POI) plays a significant role in determining the tour. A challenging question is how to transform effectively the ontological specifications into formal descriptions? Ontologies can serve as comprehensive data descriptions, including not only the names of objects and the named relations between objects but also their meaning; conceptual change raises several questions concerning the domain dependence of the POI ontology on one hand, and the universal aspects of the ontology on the other. The design and implementation of the POI ontology aims semantically to enrich the existing database. The enrichment process includes not only the addition of attributes, but foremost the addition of the assumptions about the intended meaning of the underlying vocabulary.
“Ontologies for the Design of Ecosystems.” This chapter presents guidelines and experiences of the modelling and implementation of utility ontologies, which are knowledge representations that include the general concepts required for the design of ecosystems when representing spatial and temporal data. The methods of representing the range of ontology properties are: (1) a method that includes a complete inventory of the methods of representing the properties of properties; (2) another method which shows the full range of ontology-editor alternatives with a comparative evaluation; and (3) a further method which presents an ecosystem ontology. Three utility ontologies for the general concepts of address, calendar, and landscape are presented as examples of the formalisation required during the process of ecosystem modelling. The utility ontologies and terminologies described here are the manifestation of a shared understanding and can be used as a part of the semantic grounding for the communications among web services and systems. The ontologies are implemented in the OWL 2 knowledge-representation language and a summary is provided of the tools that currently allow the user to manage this language at a high level.
“Unified Rule Approach and the Semantic Enrichment of Economic Movement Data.” This final chapter proposes a methodology for the semantic enrichment of spatial data resulting from economic behaviour. The methodology focuses on human activities in the geographic space. The semantic reference system of economic movement data is semantically enriched with an ontology based on the general analytical framework of economic evolution stated in terms of a unified rule approach. The methodology takes stock of contemporary studies of movement data and their visualisation. The goal is to provide an analytical tool that allows for the identification of their operant and generic rules. It endorses the recurring focus on moving entities and collectives, and the distinction between operational and behavioural movement patterns; however, it argues that their analysis might benefit from more explicit theoretical foundations. In this respect, the methodology shares a common starting point with spatially grounded intentional systems, which are defined as “a representational formalism and a reasoning mechanism for knowledge about an agent who acts according to changing intentions in a spatial environment.” A general way of using this knowledge for analytical purposes is suggested. The solution attempts to link ontological information with metadata as a means of providing deep meaning and context to both semantic and numeric metadata.
We conclude the preface with a description of our methodological outline for the formation of a universal ontology of the geographic space. In the development of the formal ontology, the extent of the highest level has to be limited. This presumption ensures a clearness of understanding and an ease of use. The proposed methodology consists of four interrelated parts: (1) the creation of a universal ontology of the geographic space; (2) the assessment of semantic metadata of spatial databases; (3) the assessment of semantic accordance/ similarity among concepts of various spatial data catalogues; and (4) the implementation of semantic integration of semantically harmonic concepts of spatial databases.
(1) The core methodology that enables the creation of a universal ontology of the geographic space is derived from the simultaneous treatment of social and physical reality (as in geography), which is a key element in the modelling of geographic space. Numerous features of both human (social, anthropogenic) and natural/physical origin have materialised in the geographic space. In addition, there are spatial schemes within the geographic space, such as abstract spatial formations with consequently defined boundaries, and observed spatial formations of social phenomena and physical natural phenomena. For both social and physical phenomena, the observed boundaries are either bounded (e.g. parcel or sinkhole) or continuous (e.g. population density or temperature distribution), which is reflected in the spatial concepts.
The conceptual framework of the geographic space is designed as a multilayer composition of domains consisting of heterogeneous secondary spatial concepts. Symbols and words that are commonly used for naming various spatial objects are attributes of spatial activities within the described ontology (e.g. housing—house, apartment; trading—store or warehouse). The designed ontology is independent of the features data in existing databases as well as real spatial features. The proposed methodology also follows the inherent knowledge and intuition that everyone has about the surrounding reality according to principles of naïve geography. This approach can make geographic information more easily accessible in a way that is understandable to everyone. It also has the potential to keep it easy to use.
To ensure a combination of the user’s own data and spatially enabled data that can be found on the web, a form of effective and simplified spatial data searching must be provided. This means that looking for available spatial data, for example, data on a user’s neighbourhood entities, should not explicitly require the choice of an exact (usually coded) name representing the spatial database, which (according to the user’s opinion) possibly contains the sought information. The development of a user-friendly spatial data search engine would certainly make spatially aware datasets more valuable.
The universal ontology of geographic space is a semantic reference system consisting of concepts logically organised into strict hierarchical levels. The set of concepts at the lower levels is not exhaustible but upgradeable. Combining simple concepts supports complex concepts. Some examples of the third level classes of activities are: the activity of residence, the activity of acquiring natural resources, energy, the activity of manufacturing products, leisure activities, etc. Solving the problem of the semantic interoperability of spatial data is crucial for inventing a universal semantic reference system. This problem is similar to the spatial reference system; Descartes solved the problem by inventing a coordinate system. The idea is a semantic integration system which differentiates between two independent and mutually connectable ontologies, namely a semantic reference system of geographic space (a developed universal ontology of geographic space) and an Ontology of Geo-Information Technology (OGIT).
We suggest that the information technology is presented with an OGIT, while definitions of concepts are adopted from the groups of the ISO and the OpenGIS standards. The linkage of typological spatial concepts of two semantic models (the universal ontology of geographic space and OGIT) is provided through two concepts of the universal ontology of geographic space, defined as: (a) the activity of naming geographic features with proper names; and (b) the activity of establishing geometric schemas within geographic space.
The group of subordinated concepts of a discrete object view (point, line, area), and the continuous field view of the world (mosaic cells on a grid), represent the link between the basic concepts of geographic space and the technological concepts of GIS. The semantic integration of descriptive definitions of spatial locations within coordinate spatial reference systems is based on the concept of a universal ontology of geographic space which we call the activity of naming geographic features with proper names.
Spatial data can be accessed with the help of a universal ontology of geographic space and a supportive mechanism for the semantic comparison between the available and the requested data. First, a user selects a concept (a theme) within the semantic reference system displayed, i.e. the universal ontology of geographic space that she/he is interested in. This might be, say, the concept “school”. In the next step, the user limits the search to the geographically defined administrative region (state, region and city). Consequently, the semantic part of the GIS engine processes the request and proposes a list of all available (distributed) spatial data, semantically and geographically matching the concept school. Assuming that the distributed spatial datasets offer semantic information which can be mapped by a semantic GIS engine, and assuming that the GIS—a semantic GIS engine—uses a universal ontology of geographic space, the proposed list will consist of semantically-referenced spatial data sets which describe phenomena/ objects that are the cause/consequence of the specified concept. Next, the user can read metadata about the quality of each proposed spatial data set, and can thus compare the data quality estimates and collect (buy) the one that best fits her/ his needs.
Hierarchies that tend to expand exponentially with depth can be laid out uniformly in hyperbolic space so that the distance between parents, children, and siblings is roughly the same across the hierarchy. A strict hierarchical structure of a universal ontology of geographic space is necessary in order to establish the semantic measures. Therefore, we propose the use of a hyperbolic browser (fisheye) scheme to support the visualisation and manipulation of large hierarchies of a universal ontology of geographic space especially to the non-GIS experts.
(2) The assessment of the semantic metadata of spatial databases consists of the following phases: (1) the translation of database catalogue symbols into a semantic framework – schema enrichment; (2) the assessment of semantic relations between database concepts and the symbols of a conceptual framework – the universal ontology of the geographic space; (3) the analyses of semantic suitability of the universal ontology of the geographic space for database integration; (4) the calculation of the semantic depth of the spatial database within the semantic framework; (5) the calculation of the semantic diffusion of the spatial database; and (6) the assessment of the quality of the presentation of the database semantic.
One problem of the schema enrichment (1) as means of semantic reference system is a topographic map key. Every cartographic symbol is defined with one to two expressions, but this is not enough for a complete understanding of the meaning. We need to know what mapping rules were involved, what the context of the particular symbols is—in other words, we need semantic enrichment… more words/symbols used for clearer semantics.
(3) The assessment of semantic accordance/ similarity among concepts of various spatial data catalogues is a sequence of two phases: (1) an assessment of the similarity of semantic metadata: the absolute semantic depth of databases, the relative diffusion of databases through semantic depth, the extent of the databases—diffusion); and (2) an assessment of the semantic similarity of the concepts-procedure of comparison of the database concepts.
The semantic similarity of concepts-procedure requires a special description. For the purpose of spatial database integration, it is necessary to extract the information from the database with an eye to the conceptual (thematic) detail, i.e. the semantic resolution and semantic concentration. The information on the semantic resolution and concentration of the database provides the mechanism for comparing the semantic metadata of the databases. The representation quality of the semantics for spatial data is one of the measures for estimating the development level of certain GIS. For the purpose of estimating the semantic similarity of spatial databases we have developed various measures which we consider as a metadata or partially as separate concepts held within the data. The semantic harmony of the spatial database is defined as the level of similarity of spatial databases dealing with two main parameters: the semantic resolution and the semantic concentration/dispersion of the spatial database. The greater the similarity of the semantic parameters of the databases compared, the greater the semantic harmony between them. For this purpose we established all possible permutations of the semantic relations for all concepts within the database, calculating their semantic distances, absolute semantic depth of the spatial concept, absolute semantic difference in depth, semantic depth of the unavoidable level of relation, absolute semantic difference in depth, relative semantic depth of relation, semantic depth of unavoidable level of relation. Additionally, we also defined the measures for the semantic resolution of the database, and the semantic concentration/ dispersion of the database, e.g. the semantic distance between relative concepts is the sum of segments within the semantic reference system that is required to establish the relationship between two concepts. The distance between neighbouring semantic levels has the norm value of 1.
The new analytical system was designed with an appropriate methodology for the comparison of spatial databases. The results of such uniform semantic analyses would offer a new reference layer—a comparable semantic spatial metadata layer. A uniform semantic spatial metadata layer would revive the spatial data market and enrich the global spatial data infrastructure. Simultaneously, the efficiency of data transactions by state agencies would be increased.
(4) The implementation of the semantic integration of semantically harmonic concepts of spatial databases is proposed through two approaches: (1) a local basic procedure; and (2) a procedure of semantic integration within the globally distributed spatial data infrastructure.
Some concepts of database catalogues are not expressible as concepts of a universal ontology of geographic space. In our research, we have analysed the constituting terms of the inexpressible concepts of the applicable catalogues of databases. The results of our analyses show that the most frequent term in inexpressible concepts is the term “land,” followed by the terms “building,” “area,” “construction,” and “object.” The universal ontology of the geographic space and the search engine functionality were implemented in prolog programming language. In its attempt to search a list of appropriate databases, the system may use the functionality of whichever one represents a simple semantic similarity model.
The universal ontology of geographic space is not yet available in an operational mode. We assume, however, that the coalition of visionaries will soon agree upon such an ontology, which will attract specialists to participate in the agreement to alleviate the daily tasks of professionals and the general public.
Ljubljana, June 2011Tomaž Podobnikar and Marjan Ceh