Spatial Temporal Geographic Ontology - Semantic Scholar
Spatial Temporal Geographic Ontology Zhaoqiang HUANG, Wenling XUAN, Xiuwan CHEN Institute of Remote Sensing & Geographical Information System Peking University Beijing 100871, P.R. China hzhaoq@ 126.com, [email protected], [email protected]
Abstract-Ontology is the branch of metaphysics that deals with the nature of being. Ontology is currently used by philosophers, information scientists and psychologists. A sharing ontology is required for communicating between the communicating participants. And the formal description of ontology is fundamental to data exchange standards. In recent years, ontology has been used by geographers. Many researchers have recognized the importance of geographic categories, and the relationships and interactions between geographic categories. Geographic objects are intrinsically tied to space and exist at some time. In the geographic realm, cognitive categories expose certain special features of geographic objects at surveyable scale. And that these features denote the specific ontological characteristics of geographical objects. At the same time, there has been an increasing awareness of the importance of time, and time has been an integral part of geographical information science. Thus, the research of geographic ontology must involve a fully spatiotemporal view of our world, in other words, it is about spatial temporal geographic ontology. In this paper, firstly, it reviews the recently research of geographic ontology of many experts. And it mainly explores that what is the geographic ontology and what are the research contents of geographic ontology, and how to construct an geographic ontology? The domain of geographic ontology contains objects, relations, boundaries, events, processes, qualities, and quantities of all sorts. Geographic objects are typically complex, and they will in every case have parts. An adequate ontology of geographic objects must therefore contain a theory of part and whole, or mereology. And that an adequate ontology of geographic objects must contain also a topology, a theory of boundaries and interiors, of connectedness and separation, that is integrated with a mereological theory of parts and wholes. Secondly, the paper investigates the three levels of reality: the spatial reality, the cognitive reality and geographic reality. Then it considers that geographic cognition is the key factor of geographic category. Geographic categories are related to natural, cultural, and fiat element, et al. Thirdly, in this section, the research is about an spatial temporal geographic ontology that contains space and time. The paper discusses the spatial classification and the temporal categories of geographic phenomena. Spatial objects based on
1-4244-1212-9/07/$25.00 ©2007 IEEE.
4627
object or field, which are continuants or discrete, must be represented suitably by geographic ontology. Temporal analogue of objects: there are events and processes of various kinds. Processes and events are the key elements that describe dynamic action of spatial temporal geographic objects. With the introduction of time, the requirement to take heed of the manifold interconnections between the spatial and the temporal must be proposed. The geographic phenomena can be represented as spatial or temporal, and as continuants or discrete, and as events or processes. These categories are not exist in isolation from one another. Spatial and temporal are restricted each other and represent the dynamic geographical phenomena all together. Lastly, the paper concludes that spatial temporal geographic ontology must take heed of the manifold interconnections between the spatial and the temporal and in the future a technical infrastructure be needed to provide a basis for GIS. Keywords- Spatial Temporal, Ontology, Cognitive Categories I.
Geographical Phenomena,
INTRODUCTION
Ontology is the branch of metaphysics that deals with the nature of being. And ontology is currently used by philosophers, information scientists and psychologists as a universal theory to research the being of entities and the essential of entity existing. A sharing ontology is required for communicating between the communicating participants. And the formal description of ontology is fundamental to data exchange standards. In 1984, Ontology began to be used in the domain of science and technology [1]. And there had some difference about definitions and comprehensions of ontology, but the essential of definition is the same. Studer et al. [2] proposed a definition that ontology is a conceptual and explicit and formal shared canonical description. And from the practical view, there had some constructions of ontology by researchers, but the most influence rule about constructing ontology is proposed by Gruber, such as definite and objectivity, perfectibility, consistency, the most monotony expansibility, and least constraint. In recent years, ontology has been used by geographers. Many researchers have recognized the importance of geographic categories, and the relationships and interactions between geographic categories. An ontology of geographical phenomena is needed to give a
better understanding of geographic world. Geographic objects are intrinsically tied to space and exist at some time. In the geographic realm, cognitive categories expose certain special features of geographic objects at surveyable scale. And that these features denote the specific ontological characteristics of geographical objects. At the same time, there has been an increasing awareness of the importance of time, and time has been an integral part of geographical information science. Thus, the research of geographic ontology must involve a fully spatio-temporal view of our world, in other words, it is about spatial temporal geographic ontology. In next sections, the second section reviews the recently research of geographic ontology of many experts. And it mainly explores that what is the geographic ontology and what are the research contents of geographic ontology, and how to construct an geographic ontology? The third section discusses the spatial classification and the temporal categories of geographic phenomena. The fourth section research is about an spatial temporal geographic ontology that contains space and time. And the fifth section gives the conclusion and the future work. II.
REVIEW OF THE GEOGRAPHIC ONTOLOGY RESEARCH
Geographic ontology is a new research branch of geographic information science. Geographic ontology means a set of theories and methods that the knowledge, information and data in geographic science domain are extracted to form a individual common object and compose a system in term of a certain relation, and to be conceptualized and defined definitely, and to be represented formally. Geographic kinds have the theoretical peculiarities that set them apart from kinds of other sorts. Geographic objects are not merely located in space, but are tied intrinsically to space in a manner that implies that they inherit from space many of its structural properties, such as mereological, topological, geometrical [3]. It is considered by Smith and Mark (1998) that geographical objects are typically complex, and they will in every case have parts, and so an adequate ontology of geographical objects must therefore contain a theory of part and whole, or mereology [4]. Further more, geographical objects also have boundaries, which is important as much as the constituents in the interiors. And that geographical objects are prototypically connected or contiguous, but also they are sometimes scattered or separated, and sometimes they are closed , and sometimes they are open. These above concepts are topological notions. Thus an adequate ontology of geographical objects must contain also a topology, a theory of boundaries and interiors, of connectedness and separation, that is integrated with a mereological theory of parts and wholes [3,5]. But because of geographical purposes, some certain sorts of boundary phenomena of the category of abstract boundaries can not be deal with by standard topology. For example, in splitting a land parcel to two parcels, we can not make a piece closed and another piece opened. Because abstract boundaries don't take up space, and they can be perfectly co-located one with another. So geographic ontology needs to include special mereotopological theories [3,6]. In addition, geographic
1-4244-1212-9/07/$25.00 ©2007 IEEE.
4628
ontology also considers qualitative geometry, which is the theory of concavity and convexity, of shortness and longness, of being roughly round or roughly dumbbell shaped. And it is also needed to consider the dimension of the geographic object and a general theory of spatial location [7]. Bishr et al. proposed that geographic ontology developed based on theory of topology, mereology, identify, category and dependence should facilitate modeling geospatial information of semantic consistency [8]. There are three main research of the construction method of Geographic ontology, such as the theory of ontology, humanistic investigating, and informatics. They are complement and constraint each other [9]. The research of theory of geographic ontology focuses on the construction method from the view of philosophy. Chrisman considered that it is required for remaining different opinion on real world and developing method which can integrate and transform all sorts of category systems and solve the heterogeneity of semantic of diverse category systems [10]. Kuhn investigated the semantic modeling of geographic categories from the view of conceptual integration theory in cognitive science and revealed the advantage and disadvantage of ontology which is organized by the tree structure or concept level, for instance the representative ontology WordNet. And the formalization in Haskell show that the theory of conceptual integration provides indeed a powerful paradigm for semantic modeling [11]. Smith and Mark [12] investigated a series of experiments designed to establish how non-expert subjects conceptualize geospatial phenomena and revealed considerable mismatch as between the meanings assigned to the terms 'geography' and 'geographic' by scientific geographers and by ordinary subjects. So that in fact scientific geographers are not studying geographic phenomena as such phenomena are conceptualized by naive subjects. And they proposed that geographical objects seem in contrast to be standardly conceptualized not in terms of us or function or behavior, but rather in terms of topology, geometry, location, and orientation [13]. And Mark and Turk investigated that the difference between regions and language cultures should indeed influence people the geographic concept and ontology [14]. Kuhn proposed a method to derive ontologies of geographical domains from natural language texts that describe human activities. Through its focus on actions afforded by domain objects, the method established a criterion for selecting the contents. Using an analysis of the German traffic code as a case study, kuhn demonstrated the informal parts of the process to derive such ontologies [15]. Kokla et al.[16,17] proposed a methodology for the fusion of different geographical domain ontologies with top-level ontologies, in order to provide a solid base for information exchange. The methodology uses Concept Lattices as a tool for the formalization and integration of geographical concepts and relationships encoded in different ontologies, in order to reveal their association and interaction. The domain of geographic ontology contains objects, relations, boundaries, events, processes, qualities, and quantities of all sorts.
III.
SPATIAL, COGNITIVE, AND GEOGRAPHIC REALITY
Spatial reality is like our real world, and we assume that it is existence of our real world which is populated by real objects occupying regions of space. The spatial regions populated by objects make a spatial relational system, comprising topological spatial relation, order spatial relation, metric spatial relation, and so on. And we presume objects and entities are synonymous as ontological terms. The objects should comprise things, relations, boundaries, processes, events, qualities, and quantities of all sorts in the real world. There is a domain of cognition which is directed towards spatial objects in the world. We cognize the spatial objects through the acts of sense, aesthesis, notice, idea, memory, thought, language, thinker. And interestingly, the subjects of these acts exist in a spatial domain. Cognitive Concepts work spatially in manifold fashion: through abstract models or representations of space in our minds, through a concrete being-in-space, through different sorts of mixtures of these and conceptual interaction with spatial entities [3]. Ou; cognitive capability is limited, and correspondence with the extent of cognition is limited. Our cognitive representations of space may be erroneous. Many concepts may be modified through measurement ways of experiment. Thus it is essential for cognition to establish the correspondences and interrelations among the different domains of spatial entities and relation in spatial and cognitive domain. Geographic reality is a sort of region partition in which geographical phenomena happen. Geographic reality includes many different sorts of properties and entities, which can be dealt with theoretically in many different methods. In the mesoscopic of spatial reality, through our cognition, we can understand straightforwardly such as rivers, bridges and forests, and on the other hand we can cognize the geographic objects like bays induced by human cognition and like nations and boundaries which are fiat. So geographic categories are related to natural, cultural, and fiat element, et al. IV.
SPATIAL-TEMPORAL PHENOMENA
A.
Spatial Classification and Temporal Categories In general, we consider the reality world with various kinds of objects, such as river, bridge, house, desert, the 10.00 a.m. train from Beijing to Tianjin on 2nd March 2007, Beijing city, and so on. These objects comprise a wide range of degrees of materiality, of naturalness, of definiteness, and of stability. Some objects have spatial parts, and exist as an entirety at each moment of its history (e.g., houses, roads, cities, and bridge). And in philosophical terminology or from ontological perspective these spatial objects are continuants [18]. It is based on object-oriented so that the world is often taken to be the totality of things. But in world there are phenomena such as remove animals, artifacts, and the trees, and infinitely various but not cleanly divided rock outcropping, cloud formation, lichen, and so on. Thus we need to use the field view seriously for describing these non-objects feature of reality-it is not just a relict of the days when the most sophisticated computer output device was the line-printer so
1-4244-1212-9/07/$25.00 ©2007 IEEE.
4629
that it was natural to handle geographical data in raster format [19]. A spatial field is a set of values mapped from spatial locations. The field changes through such examples as temperature, pressure, humidity, wind-speed, soil type, population density, and inhabitation, as well as more recondite things such as 'distance to the nearest highway', and all of which can be taken as field values. Like object, field changes over time. Nowadays, it is generally recognized that object and field views are necessary for spatial reality world. So that we use the idea of an 'object-field', for instance a field whose values are objects, to handle the field-like 'planar enforced' structures variously referred to as adjacent polygon models or partitions, such as transportation application. Temporal domain has already been included in spatial reality world. The temporal cases are analogical with spatial cases [20]. In time domain, there are events and processes of various kinds. The events have a relatively sharply delineated location in space and time, such as the sinking of the Titanic, the men 100m final of Olympic Games in 2004. But there are also numerous happenings which are ill-defined, such as the last night's rain, a house repair job, road construction project, urban expansion, and real-estate transition. And these things have temporal parts and do not exist in their entirety at any one time, and happen and are then gone. So in philosophical terminology or from ontological perspective these happenings are occurrents. Now there is no consensus as to the exact meaning of process, in part because it is used in various ways. The process may include several actions and objects during the time interval from beginning to ending, for instance the process of making a pot of tea, which includes the heating the water, warming the pot, putting in the tea-leaves, and pouring the boiling water on top of them. And the process can include uniform states of change, e.g., rolling, getting warmer, colder, denser, darker, and homogeneous aggregates of many small, relatively featureless events, e.g., many individual drops falling, many steps walking, and heterogeneous aggregates of larger, structured events, e.g., playing basketball [18]. For event and object view, the process can be consisted of several events and objects.
B.
General Spatial-Temporal Geographic Things Spatial and temporal are restricted each other and represent the dynamic geographical phenomena all together. The geographic phenomena being in the world have the spatiotemporal features. Or we can say that there are spatio-temporal locations: portions of space-time for geographic phenomena. Objects/fields and events/processes represent the spatial and temporal respectively, and they interact with each other. We can consider the general phenomena of the process that Tom gets up and goes to work. One day, Tom gets up at 6:30 a.m., and at 7:00 has breakfast, at 7:30 goes to company, and at 8:00 works on some documents. For this process, Tom is the subjective object, and getting up at 6:30 a.m. is an event enduring a short time as instant, and having breakfast is an event enduring a interval time, going to company also can be made as an event or a process, and working on some documents can be taken as an event or a process. From object view, we can say there are several status of Tom in different
space and time. From event view, we can say there are several events happening on Tom. So the geographical phenomena is a spatio-temporal entity of which the object- and event-view are both partial in. Many geographical phenomena resemble this example to some extent. Using various different views of all such spatio-temporal phenomena, it is significant to tracing the sequence of spatial location affected by them, or the sequence of events associated with the phenomenon at one spatial location. Thus Spatial and temporal features are the inherent characteristic of the general geographical phenomena. The integrity of geographical phenomena is considered to explicitly represent dynamic spatio-temporal semantic.
[8]
[9]
[10]
[11]
[12]
[13]
v.
CONCLUSION
This paper discusses the feature of geographical ontology and spatial temporal ontology. And it proposed to use various different views to analysis dynamic geographic phenomena. In the future, we need to build a technical infrastructure as a basis for implementing the representation and reasoning.
[14]
[15] REFERENCES [1] [2] [3]
[4] [5] [6]
[7]
SHI Zhongzhi, Translated term of ontology[ J]. Chinese Science and Technology Terms Journal, 2004,6(4):13-14. R Studer et al. Knowledge Engineering Principles and Methods. IEEE Transactions on Data and Knowledge Engineering. 25(1-2): 161-197 Barry Smith and David M. Mark. Ontology and Geographic Kinds. Proceedings, International Symposium on Spatial Data Handling (SDH'98), Vancouver, Canada, 12-15 July, 1998, pp. 308-320. Simons, P. M., 1987. Parts. An Essay in Ontology. Oxford: Clarendon Press. Smith, B., 1996. Mereotopology: A Theory of Parts and Boundaries. Data and Knowledge Engineering, 20, 287-303. Smith, B., and Varzi, A., 1997. Fiat and Bona Fide Boundaries: An Essay on the Foundations of Geography, in S. C. Hirtle and A. U. Frank (eds.), Spatial Information Theory. International Conference COSIT '97. Laurel Highlands, Pennsylvania, October 1997 (Lecture Notes in Computer Science 1329), Berlin/New York: Springer Verlag, 103-119. Casati, R., Smith, B., Varzi, A., Ontological tools for geographic representation. Guarino, N., Proceedings of FOIS '98, Trento, Italy, 1998. Amsterdam, Netherlands:IOS Press, 1998:77-85.
1-4244-1212-9/07/$25.00 ©2007 IEEE.
4630
[16]
[17]
[18]
[19] [20]
Bishr, Y., Kuhn, W., Ontology-based modeling of geospatial information. Proceedings of 3rd AGILE conference on geographical information science, Helsinki/Espoo, Finland,2000. Mark, D. M., Egenhorfer, M. J., Hirtle, S., et al. Ontological foundations of geographic information science. UCGIS emerging research theme technical report, 2000. Chrisman, N., Building GIS without foundations: ontology from a social practice perspective. Proceedings of GISicence, Savannah, Georgia,USA, 2000. Kuhn, W., Modeling the semantics of geographic categories through conceptual integration. Proceedings of GIScience, Boulder, Colorado, USA, 2002. Smith, B., Mark, D. M., Geographical Categories: an Ontological Investigation. International Journal of Geographical Information Science, 2001,15(7):591-612. Mark, D. M., Smith, B., and Tversky, B., Ontology and Geographic Objects: An Empirical Study of Cognitive Categorization. Freksa, C., Mark, D. M. , Spatial information theory: a theoretical basis for GIS, lecture notes in computer science. Berlin: Springer-Verlag, 1999:283298. Mark, D. M., Turk, A. G., Landscape Categories in Yindjibarndi:Ontology, Environment, and Language. Kuhn, W., Worboys, M. F., Timpf, s. Spatial information theory: foundation of geographic information science. Berlin: Springer-Verlag, Lecture Notes in Computer Science. No.2825,2003:28-45. Kuhn, Werner, Ontologies in support of activities in geographical space. International Journal of Geographical Information Science, 2001,15(7): 613-631. Kokla, Margarita, Kavouras, Marinos. Fusion of top-level and geographical domain ontologies based on context formation and complementarity. International Journal of Geographical Information Science, 2001,15(7): 679-687. Kavouras, Marinos, Kokla, Margarita. A method for the formalization and integration of geographical categorizations. International Journal of Geographical Information Science, 2002, 16(5):439-453 Antony Galton. Desiderata for a Spatio-temporal Geo-ontology. Kuhn, W., Worboys, M. F., Timpf, s. Spatial information theory: foundation of geographic information science. Berlin: Springer-Verlag, Lecture Notes in Computer Science. No.2825,2003:1-12. Donna J. Peuquet, Making space for time: issues in space-time data representation. GeoInformatica, 2001,5(1):11-32. Michael F. Worboys, Unifying the spatial and temporal components of geographic information. In T. C. Waugh and R. G. Healey, editors, Advances in Geographic Information Systems: Proceedings of the Sixth International Symposium on Spatial Data Handling, Edinburgh, pages 505-517, London, 1994. Taylor and Francis.
Abstract-Ontology is the branch of metaphysics that deals with the nature of being. Ontology is currently used by philosophers, information scientists and psychologists. A sharing ontology is required for communicating between the communicating participants. And the formal description of ontology is fundamental to data exchange standards. In recent years, ontology has been used by geographers. Many researchers have recognized the importance of geographic categories, and the relationships and interactions between geographic categories. Geographic objects are intrinsically tied to space and exist at some time. In the geographic realm, cognitive categories expose certain special features of geographic objects at surveyable scale. And that these features denote the specific ontological characteristics of geographical objects. At the same time, there has been an increasing awareness of the importance of time, and time has been an integral part of geographical information science. Thus, the research of geographic ontology must involve a fully spatiotemporal view of our world, in other words, it is about spatial temporal geographic ontology. In this paper, firstly, it reviews the recently research of geographic ontology of many experts. And it mainly explores that what is the geographic ontology and what are the research contents of geographic ontology, and how to construct an geographic ontology? The domain of geographic ontology contains objects, relations, boundaries, events, processes, qualities, and quantities of all sorts. Geographic objects are typically complex, and they will in every case have parts. An adequate ontology of geographic objects must therefore contain a theory of part and whole, or mereology. And that an adequate ontology of geographic objects must contain also a topology, a theory of boundaries and interiors, of connectedness and separation, that is integrated with a mereological theory of parts and wholes. Secondly, the paper investigates the three levels of reality: the spatial reality, the cognitive reality and geographic reality. Then it considers that geographic cognition is the key factor of geographic category. Geographic categories are related to natural, cultural, and fiat element, et al. Thirdly, in this section, the research is about an spatial temporal geographic ontology that contains space and time. The paper discusses the spatial classification and the temporal categories of geographic phenomena. Spatial objects based on
1-4244-1212-9/07/$25.00 ©2007 IEEE.
4627
object or field, which are continuants or discrete, must be represented suitably by geographic ontology. Temporal analogue of objects: there are events and processes of various kinds. Processes and events are the key elements that describe dynamic action of spatial temporal geographic objects. With the introduction of time, the requirement to take heed of the manifold interconnections between the spatial and the temporal must be proposed. The geographic phenomena can be represented as spatial or temporal, and as continuants or discrete, and as events or processes. These categories are not exist in isolation from one another. Spatial and temporal are restricted each other and represent the dynamic geographical phenomena all together. Lastly, the paper concludes that spatial temporal geographic ontology must take heed of the manifold interconnections between the spatial and the temporal and in the future a technical infrastructure be needed to provide a basis for GIS. Keywords- Spatial Temporal, Ontology, Cognitive Categories I.
Geographical Phenomena,
INTRODUCTION
Ontology is the branch of metaphysics that deals with the nature of being. And ontology is currently used by philosophers, information scientists and psychologists as a universal theory to research the being of entities and the essential of entity existing. A sharing ontology is required for communicating between the communicating participants. And the formal description of ontology is fundamental to data exchange standards. In 1984, Ontology began to be used in the domain of science and technology [1]. And there had some difference about definitions and comprehensions of ontology, but the essential of definition is the same. Studer et al. [2] proposed a definition that ontology is a conceptual and explicit and formal shared canonical description. And from the practical view, there had some constructions of ontology by researchers, but the most influence rule about constructing ontology is proposed by Gruber, such as definite and objectivity, perfectibility, consistency, the most monotony expansibility, and least constraint. In recent years, ontology has been used by geographers. Many researchers have recognized the importance of geographic categories, and the relationships and interactions between geographic categories. An ontology of geographical phenomena is needed to give a
better understanding of geographic world. Geographic objects are intrinsically tied to space and exist at some time. In the geographic realm, cognitive categories expose certain special features of geographic objects at surveyable scale. And that these features denote the specific ontological characteristics of geographical objects. At the same time, there has been an increasing awareness of the importance of time, and time has been an integral part of geographical information science. Thus, the research of geographic ontology must involve a fully spatio-temporal view of our world, in other words, it is about spatial temporal geographic ontology. In next sections, the second section reviews the recently research of geographic ontology of many experts. And it mainly explores that what is the geographic ontology and what are the research contents of geographic ontology, and how to construct an geographic ontology? The third section discusses the spatial classification and the temporal categories of geographic phenomena. The fourth section research is about an spatial temporal geographic ontology that contains space and time. And the fifth section gives the conclusion and the future work. II.
REVIEW OF THE GEOGRAPHIC ONTOLOGY RESEARCH
Geographic ontology is a new research branch of geographic information science. Geographic ontology means a set of theories and methods that the knowledge, information and data in geographic science domain are extracted to form a individual common object and compose a system in term of a certain relation, and to be conceptualized and defined definitely, and to be represented formally. Geographic kinds have the theoretical peculiarities that set them apart from kinds of other sorts. Geographic objects are not merely located in space, but are tied intrinsically to space in a manner that implies that they inherit from space many of its structural properties, such as mereological, topological, geometrical [3]. It is considered by Smith and Mark (1998) that geographical objects are typically complex, and they will in every case have parts, and so an adequate ontology of geographical objects must therefore contain a theory of part and whole, or mereology [4]. Further more, geographical objects also have boundaries, which is important as much as the constituents in the interiors. And that geographical objects are prototypically connected or contiguous, but also they are sometimes scattered or separated, and sometimes they are closed , and sometimes they are open. These above concepts are topological notions. Thus an adequate ontology of geographical objects must contain also a topology, a theory of boundaries and interiors, of connectedness and separation, that is integrated with a mereological theory of parts and wholes [3,5]. But because of geographical purposes, some certain sorts of boundary phenomena of the category of abstract boundaries can not be deal with by standard topology. For example, in splitting a land parcel to two parcels, we can not make a piece closed and another piece opened. Because abstract boundaries don't take up space, and they can be perfectly co-located one with another. So geographic ontology needs to include special mereotopological theories [3,6]. In addition, geographic
1-4244-1212-9/07/$25.00 ©2007 IEEE.
4628
ontology also considers qualitative geometry, which is the theory of concavity and convexity, of shortness and longness, of being roughly round or roughly dumbbell shaped. And it is also needed to consider the dimension of the geographic object and a general theory of spatial location [7]. Bishr et al. proposed that geographic ontology developed based on theory of topology, mereology, identify, category and dependence should facilitate modeling geospatial information of semantic consistency [8]. There are three main research of the construction method of Geographic ontology, such as the theory of ontology, humanistic investigating, and informatics. They are complement and constraint each other [9]. The research of theory of geographic ontology focuses on the construction method from the view of philosophy. Chrisman considered that it is required for remaining different opinion on real world and developing method which can integrate and transform all sorts of category systems and solve the heterogeneity of semantic of diverse category systems [10]. Kuhn investigated the semantic modeling of geographic categories from the view of conceptual integration theory in cognitive science and revealed the advantage and disadvantage of ontology which is organized by the tree structure or concept level, for instance the representative ontology WordNet. And the formalization in Haskell show that the theory of conceptual integration provides indeed a powerful paradigm for semantic modeling [11]. Smith and Mark [12] investigated a series of experiments designed to establish how non-expert subjects conceptualize geospatial phenomena and revealed considerable mismatch as between the meanings assigned to the terms 'geography' and 'geographic' by scientific geographers and by ordinary subjects. So that in fact scientific geographers are not studying geographic phenomena as such phenomena are conceptualized by naive subjects. And they proposed that geographical objects seem in contrast to be standardly conceptualized not in terms of us or function or behavior, but rather in terms of topology, geometry, location, and orientation [13]. And Mark and Turk investigated that the difference between regions and language cultures should indeed influence people the geographic concept and ontology [14]. Kuhn proposed a method to derive ontologies of geographical domains from natural language texts that describe human activities. Through its focus on actions afforded by domain objects, the method established a criterion for selecting the contents. Using an analysis of the German traffic code as a case study, kuhn demonstrated the informal parts of the process to derive such ontologies [15]. Kokla et al.[16,17] proposed a methodology for the fusion of different geographical domain ontologies with top-level ontologies, in order to provide a solid base for information exchange. The methodology uses Concept Lattices as a tool for the formalization and integration of geographical concepts and relationships encoded in different ontologies, in order to reveal their association and interaction. The domain of geographic ontology contains objects, relations, boundaries, events, processes, qualities, and quantities of all sorts.
III.
SPATIAL, COGNITIVE, AND GEOGRAPHIC REALITY
Spatial reality is like our real world, and we assume that it is existence of our real world which is populated by real objects occupying regions of space. The spatial regions populated by objects make a spatial relational system, comprising topological spatial relation, order spatial relation, metric spatial relation, and so on. And we presume objects and entities are synonymous as ontological terms. The objects should comprise things, relations, boundaries, processes, events, qualities, and quantities of all sorts in the real world. There is a domain of cognition which is directed towards spatial objects in the world. We cognize the spatial objects through the acts of sense, aesthesis, notice, idea, memory, thought, language, thinker. And interestingly, the subjects of these acts exist in a spatial domain. Cognitive Concepts work spatially in manifold fashion: through abstract models or representations of space in our minds, through a concrete being-in-space, through different sorts of mixtures of these and conceptual interaction with spatial entities [3]. Ou; cognitive capability is limited, and correspondence with the extent of cognition is limited. Our cognitive representations of space may be erroneous. Many concepts may be modified through measurement ways of experiment. Thus it is essential for cognition to establish the correspondences and interrelations among the different domains of spatial entities and relation in spatial and cognitive domain. Geographic reality is a sort of region partition in which geographical phenomena happen. Geographic reality includes many different sorts of properties and entities, which can be dealt with theoretically in many different methods. In the mesoscopic of spatial reality, through our cognition, we can understand straightforwardly such as rivers, bridges and forests, and on the other hand we can cognize the geographic objects like bays induced by human cognition and like nations and boundaries which are fiat. So geographic categories are related to natural, cultural, and fiat element, et al. IV.
SPATIAL-TEMPORAL PHENOMENA
A.
Spatial Classification and Temporal Categories In general, we consider the reality world with various kinds of objects, such as river, bridge, house, desert, the 10.00 a.m. train from Beijing to Tianjin on 2nd March 2007, Beijing city, and so on. These objects comprise a wide range of degrees of materiality, of naturalness, of definiteness, and of stability. Some objects have spatial parts, and exist as an entirety at each moment of its history (e.g., houses, roads, cities, and bridge). And in philosophical terminology or from ontological perspective these spatial objects are continuants [18]. It is based on object-oriented so that the world is often taken to be the totality of things. But in world there are phenomena such as remove animals, artifacts, and the trees, and infinitely various but not cleanly divided rock outcropping, cloud formation, lichen, and so on. Thus we need to use the field view seriously for describing these non-objects feature of reality-it is not just a relict of the days when the most sophisticated computer output device was the line-printer so
1-4244-1212-9/07/$25.00 ©2007 IEEE.
4629
that it was natural to handle geographical data in raster format [19]. A spatial field is a set of values mapped from spatial locations. The field changes through such examples as temperature, pressure, humidity, wind-speed, soil type, population density, and inhabitation, as well as more recondite things such as 'distance to the nearest highway', and all of which can be taken as field values. Like object, field changes over time. Nowadays, it is generally recognized that object and field views are necessary for spatial reality world. So that we use the idea of an 'object-field', for instance a field whose values are objects, to handle the field-like 'planar enforced' structures variously referred to as adjacent polygon models or partitions, such as transportation application. Temporal domain has already been included in spatial reality world. The temporal cases are analogical with spatial cases [20]. In time domain, there are events and processes of various kinds. The events have a relatively sharply delineated location in space and time, such as the sinking of the Titanic, the men 100m final of Olympic Games in 2004. But there are also numerous happenings which are ill-defined, such as the last night's rain, a house repair job, road construction project, urban expansion, and real-estate transition. And these things have temporal parts and do not exist in their entirety at any one time, and happen and are then gone. So in philosophical terminology or from ontological perspective these happenings are occurrents. Now there is no consensus as to the exact meaning of process, in part because it is used in various ways. The process may include several actions and objects during the time interval from beginning to ending, for instance the process of making a pot of tea, which includes the heating the water, warming the pot, putting in the tea-leaves, and pouring the boiling water on top of them. And the process can include uniform states of change, e.g., rolling, getting warmer, colder, denser, darker, and homogeneous aggregates of many small, relatively featureless events, e.g., many individual drops falling, many steps walking, and heterogeneous aggregates of larger, structured events, e.g., playing basketball [18]. For event and object view, the process can be consisted of several events and objects.
B.
General Spatial-Temporal Geographic Things Spatial and temporal are restricted each other and represent the dynamic geographical phenomena all together. The geographic phenomena being in the world have the spatiotemporal features. Or we can say that there are spatio-temporal locations: portions of space-time for geographic phenomena. Objects/fields and events/processes represent the spatial and temporal respectively, and they interact with each other. We can consider the general phenomena of the process that Tom gets up and goes to work. One day, Tom gets up at 6:30 a.m., and at 7:00 has breakfast, at 7:30 goes to company, and at 8:00 works on some documents. For this process, Tom is the subjective object, and getting up at 6:30 a.m. is an event enduring a short time as instant, and having breakfast is an event enduring a interval time, going to company also can be made as an event or a process, and working on some documents can be taken as an event or a process. From object view, we can say there are several status of Tom in different
space and time. From event view, we can say there are several events happening on Tom. So the geographical phenomena is a spatio-temporal entity of which the object- and event-view are both partial in. Many geographical phenomena resemble this example to some extent. Using various different views of all such spatio-temporal phenomena, it is significant to tracing the sequence of spatial location affected by them, or the sequence of events associated with the phenomenon at one spatial location. Thus Spatial and temporal features are the inherent characteristic of the general geographical phenomena. The integrity of geographical phenomena is considered to explicitly represent dynamic spatio-temporal semantic.
[8]
[9]
[10]
[11]
[12]
[13]
v.
CONCLUSION
This paper discusses the feature of geographical ontology and spatial temporal ontology. And it proposed to use various different views to analysis dynamic geographic phenomena. In the future, we need to build a technical infrastructure as a basis for implementing the representation and reasoning.
[14]
[15] REFERENCES [1] [2] [3]
[4] [5] [6]
[7]
SHI Zhongzhi, Translated term of ontology[ J]. Chinese Science and Technology Terms Journal, 2004,6(4):13-14. R Studer et al. Knowledge Engineering Principles and Methods. IEEE Transactions on Data and Knowledge Engineering. 25(1-2): 161-197 Barry Smith and David M. Mark. Ontology and Geographic Kinds. Proceedings, International Symposium on Spatial Data Handling (SDH'98), Vancouver, Canada, 12-15 July, 1998, pp. 308-320. Simons, P. M., 1987. Parts. An Essay in Ontology. Oxford: Clarendon Press. Smith, B., 1996. Mereotopology: A Theory of Parts and Boundaries. Data and Knowledge Engineering, 20, 287-303. Smith, B., and Varzi, A., 1997. Fiat and Bona Fide Boundaries: An Essay on the Foundations of Geography, in S. C. Hirtle and A. U. Frank (eds.), Spatial Information Theory. International Conference COSIT '97. Laurel Highlands, Pennsylvania, October 1997 (Lecture Notes in Computer Science 1329), Berlin/New York: Springer Verlag, 103-119. Casati, R., Smith, B., Varzi, A., Ontological tools for geographic representation. Guarino, N., Proceedings of FOIS '98, Trento, Italy, 1998. Amsterdam, Netherlands:IOS Press, 1998:77-85.
1-4244-1212-9/07/$25.00 ©2007 IEEE.
4630
[16]
[17]
[18]
[19] [20]
Bishr, Y., Kuhn, W., Ontology-based modeling of geospatial information. Proceedings of 3rd AGILE conference on geographical information science, Helsinki/Espoo, Finland,2000. Mark, D. M., Egenhorfer, M. J., Hirtle, S., et al. Ontological foundations of geographic information science. UCGIS emerging research theme technical report, 2000. Chrisman, N., Building GIS without foundations: ontology from a social practice perspective. Proceedings of GISicence, Savannah, Georgia,USA, 2000. Kuhn, W., Modeling the semantics of geographic categories through conceptual integration. Proceedings of GIScience, Boulder, Colorado, USA, 2002. Smith, B., Mark, D. M., Geographical Categories: an Ontological Investigation. International Journal of Geographical Information Science, 2001,15(7):591-612. Mark, D. M., Smith, B., and Tversky, B., Ontology and Geographic Objects: An Empirical Study of Cognitive Categorization. Freksa, C., Mark, D. M. , Spatial information theory: a theoretical basis for GIS, lecture notes in computer science. Berlin: Springer-Verlag, 1999:283298. Mark, D. M., Turk, A. G., Landscape Categories in Yindjibarndi:Ontology, Environment, and Language. Kuhn, W., Worboys, M. F., Timpf, s. Spatial information theory: foundation of geographic information science. Berlin: Springer-Verlag, Lecture Notes in Computer Science. No.2825,2003:28-45. Kuhn, Werner, Ontologies in support of activities in geographical space. International Journal of Geographical Information Science, 2001,15(7): 613-631. Kokla, Margarita, Kavouras, Marinos. Fusion of top-level and geographical domain ontologies based on context formation and complementarity. International Journal of Geographical Information Science, 2001,15(7): 679-687. Kavouras, Marinos, Kokla, Margarita. A method for the formalization and integration of geographical categorizations. International Journal of Geographical Information Science, 2002, 16(5):439-453 Antony Galton. Desiderata for a Spatio-temporal Geo-ontology. Kuhn, W., Worboys, M. F., Timpf, s. Spatial information theory: foundation of geographic information science. Berlin: Springer-Verlag, Lecture Notes in Computer Science. No.2825,2003:1-12. Donna J. Peuquet, Making space for time: issues in space-time data representation. GeoInformatica, 2001,5(1):11-32. Michael F. Worboys, Unifying the spatial and temporal components of geographic information. In T. C. Waugh and R. G. Healey, editors, Advances in Geographic Information Systems: Proceedings of the Sixth International Symposium on Spatial Data Handling, Edinburgh, pages 505-517, London, 1994. Taylor and Francis.
