Research Group Geoinformation


Home | Research | Publications | Download | Projects | Teaching | Staff



... to the Cartography Group

Bittner Steffen

An agent based Model of Reality in a Cadastre

Motivation.

"Land, after all, is the ultimate resource from which all wealth comes." [3,p.21]

The foundation for human settlement is the land without which human life would be impossible. Its value and importance is therefore extremely high. Land cannot be increased, it is a limited resource and an important factor of production. Much effort is invested into the management of land. Cadastral systems are developed all over the world for this purpose. Consequently a major demand for effective and efficient organization of these systems exists [3].

Foundation for efficient cadastral systems is the understanding of the reality which the system should correctly represent. It is not sufficient to investigate only the cadastral registry with its content and input and output operations. The registration process in the cadastral registry captures only a part of reality. The complexity of phenomena involved makes it necessary to widen the scope to the more general view of reality in a cadastre that comprises the cadastral registry as well as people acting in the real world. This allows to represent a more comprehensive view on the cadastral domain.

Reality in a cadastre is determined by a broad variety of phenomena. On one side there are physical phenomena, for instance, land pieces and human being acting on land. On the other side there are social constructs, for instance, ownership, legal persons and rights. These social constructs influence the behaviour of people in physical reality and play thus a crucial role for the representation. A model representing reality in a cadastre has to be a model of social reality.

Much research has been done to understand physical and social reality around us. There is only one reality and the question arises how they are connected. How is the fact that a person is the owner of a parcel based on the physical properties of human beings and parts of the earth surface? The improvement of the understanding of the small part of reality, namely reality in a cadastre, contributes to our understanding of reality in general.

One issue in the analysis of social reality is the complexity of the involved phenomena and the difficulty to extract components small enough to construct models. Cadastre is an appropriate field to explore social reality because it is on on side a relatively separated part of reality with limited influences from and to other parts of the real world. On the other side the rules structuring social reality in a cadastre are to a high degree known and represented in laws and thus the field is open to an analysis.

There are tools necessary to support the improvement of our understanding of reality and its consequences for the construction of more efficient cadastral systems. Computational models are appropriate for this purpose because they allow the simulation of processes in reality. For instance a model representing reality in a cadastre enables the simulation of the behaviour of a person violating the legal rules and its consequences to the cadastral registry.

It is advantageous to use a formal tool for the model of reality in a cadastre, i.e. a formal specification language with clear semantics. This allows to represent the domain in a clear and unambiguous way and helps to avoid misunderstandings. The formal model should be executable because this enables the construction of computational models. Computational models based on a formal executable specification have advantages compared with models not based on a formal foundation. A model based on a formal specification formally correct represents its specification, i.e. it does the right things with respect to the specification. This enables the testing of the model with appropriate test cases to investigate its correspondence to the reality it represents.

Research Question.

The key issue for the analysis of reality in a cadastre is the understanding of the social concepts involved. The question is how to represent social reality. Formal models have advantages compared with non formal approaches. The main research question of this thesis is therefore:



Is it possible to construct a formal, computational model of (social) reality in a cadastre?

Hypothesis.

Social reality in a cadastre is highly determined by institutional concepts (e.g. ownership, parcel,...). In the model it is sufficient to represent only institutional reality. The relevant institutional concepts are defined in laws (especially the cadastral law). By the analysis of the cadastral law it is possible to extract the rules constituting institutional reality. Searle's theory on institutional reality [11] introduces an approach dealing with institutional concepts and gives the theoretical and philosophical background of this work. Social reality in general and institutional reality in particular are determined by the behaviour of human beings. Thus for the model a representation of human intentions and behaviour is necessary. Agent theory [9,5,13] gives the conceptual framework for this task. It allows formally representing human intentions and behaviour necessary for the model of reality in a cadastre by agents. The hypothesis of this thesis reads as follows:

A formal, computational model of institutional reality represented in an agent based framework represents the relevant aspects of reality in a cadastre.

Approach.

The basic approach of this thesis is the construction of a formal and computational model of reality in a cadastre. The correctness of the model will verified by agent based simulation of appropriate and characteristic processes from the real world in the constructed model.

The following tasks characterize the approach of this thesis:
Start on a sophisticated philosophical foundation with and analysis of the domain in question (modeling of institutional reality in a cadastre).
Use of agents and agent theory as conceptual framework to represent the domain.
Use of an algebraic specification [7,4,6] as formal framework to represent the domain.
Implementation of the algebraic specification in the functional programming language Haskell. [12,10,1] to reach a computational model.
Empirical validation of the model with appropriate case studies.

The thesis applies Searle's theory of institutional reality to the analysis of reality in a cadastre. The analysis results in a basic ontology of reality in a cadastre. Based on this ontology a model will be developed in an agent based framework. This model will be represented with a Haskell specification. This specification allows agent based simulation of processes in reality. To verify the correctness of the model two case studies will be performed. One case study is the transfer of ownership of a parcel between two persons and the second is a conflict regarding land use and the resulting legal action and judgement execution.

Goal of the thesis.

The general goal of this thesis is the investigation of the structure of reality in a cadastre as part of social reality. It comprises the development a computational model of this reality. Based on this model expressible case studies will be developed that represent a comprehensive part of phenomena determining reality in a cadastre. These case studies are the transfer of ownership on a parcel and conflicts regarding the use of a piece of land between the owner, the authorized user of the land and an unauthorized user of the land. The conflict leads to a legal action by the owner and a judgement execution to end the unauthorized land use. The result of these case studies are simulations of social processes in a cadastre.

This thesis applies theoretical results from the field of Philosophy and Artificial Intelligence to the cadastral domain. It focuses on the Austrian cadastral system [8] to achieve a more realistic formalization. It concentrates on the legal part of the cadastre, omitting spatial issues because the questions raised by the research question are mainly situated in this field. On the philosophical and formal foundation social processes will be simulated according to the cases studies.

Goal of this work is to develop a content theory in opposition to a mechanism theory [2]. It is mostly interested in reality in the cadastral domain and not so much in the representation mechanism. The focus is directed to the analysis of reality and a conceptualization of the phenomena influencing reality. I does not develop a formalism for agent based simulation. It uses a representation mechanism to apply the conceptual framework developed.

Expected achievements.

The expected result of this thesis is that it is possible to construct a model of reality in a cadastre that can be successfully validated with the case studies. The development of a computational model in an executable specification language allows the validation by testing the specification. A realistic formalization will be achieved by the orientation on the concrete Austrian system. It will be expected that the analysis of reality in a cadastre leads to a general ontology of the cadastral domain which is applicable to other legal systems. The agent based framework to developed in this thesis will be extensible to larger parts of the Austrian cadastral law as well as other legal domains possibly from other legal systems. From the choice of the functional programming language we expect a clear and understandable representation that helps to avoid mistakes and support the expression of the ideas of this thesis.

The scientific contribution of a formal model model of reality in a cadastre has a practical and a theoretical dimension. It improves the understanding of reality in a cadastre and supports the construction of efficient cadastral systems on the practical side as well as a more realistic view of the world around us on the theoretical side.


References

[1]
R. Bird and P. Wadler. Introduction to Functional Programming. Prentice Hall International, 1988.
[2]
B. Chandrasekaran, J. Josephson, and V. R. Benjamins. What are ontologies, and why do we need them? IEEE Intelligent Systems, 14(1):20-26, 1999.
[3]
P. F. Dale and J. D. McLaughlin. Land Information Management, An introduction with special reference to cadastral systems in Third World countries. Oxford University Press, Oxford, 1989.
[4]
H.-D. Ehrlich, M. Gogolla, and U. W. Lipeck. Algebraisch Spezifikation abstrakter Datentypen. B.G.Teubner, Stuttgart, 1989.
[5]
J. Ferber. Multi-Agent Systems. An introduction to Distributed Artificial Intelligence. Addison-Wesley, 1999.
[6]
I. V. Horebeek and J. Levi. Algebraic Specification in Software Engineering. Springer-Verlag, 1989.
[7]
J. Loeckx, H.-D. Ehrlich, and M. Wolf. Specification of Abstract Data Types. John Wiley and Sons, B.G. Teubner, Chichester, New York, Brisbane, Toronto, Singapore, Stuttgart, Leipzig, 1996.
[8]
K.-H. Marent and G. Preisl. Grundbuchsrecht. Linde Verlag, Wien, 1994.
[9]
G. O'Hare and N. Jennnings, editors. Foundations of Distributed Artificial Intelligence. Sixth-Generation Computer Technology Series. John Wiley and Sons, Inc., 1996.
[10]
J. H. F. Paul Hudak, John Peterson. A gentle introduction to haskell. 1997.
[11]
J. Searle. The construction of social reality. The Free Press, New York, 1995.
[12]
S. Thompson. The Craft of Functional Programming. Addison-Wesley, 1996.
[13]
M. Wooldridge and N. R. Jennings. Intelligent agents: Theory and practice. Knowledge Engineering Review, 10(2), 1995.

Powered by CMSimple