Conference Programme: 6th Fuzzy Days in Dortmund [LONG]

Stephan Lehmke (Stephan.Lehmke@cs.uni-dortmund.de)
Tue, 27 Apr 1999 18:40:00 +0200 (MET DST)

(apologies if you receive this announcement more than once)

Welcome

Fuzzy Days in Dortmund were held for the first time in 1991. Initially, the
conference was intended for scientists and practitioners as a platform for
discussion on the theory and application of fuzzy logic. Early on, synergetic
links with neural networks were included and the conference evolved gradually
to embrace the full spectrum of what is now called Computational Intelligence
(CI). Therefore it seemed logical to launch the 4th Fuzzy Days in 1994 as a
conference for CI - one of the world's first conferences featuring fuzzy logic,
neural networks and evolutionary algorithms together in one event. Following
this successful tradition, the 6th Fuzzy Days' aim is to provide an international
forum for reporting significant results on the theory and application of
CI-methods.

Again, ITG, GMM, GMA and GI, the well-known German societies which
represent a considerable number of engineers and computer scientists, are
co-organising and supporting the conference. Further support is given by a
multitude of German and international institutions.

We have received once again a remarkable number of proposals. I would like to
express my gratitude to all who have been interested in presenting their work
and to the members of the programme committee for their valuable work (in
this edition each paper was reviewed by five referees). In particular, I wish to
thank all keynote and tutorial speakers for their commitment. Last but not
least, I am obliged to Deutsche Forschungsgemeinschaft and
Kommunalverband Ruhrgebiet for their financial support.

We are confident that the 6th Fuzzy Days will be an excellent forum to
exchange ideas, share experiences, present results and discuss the potentials of
CI. I look forward to meeting you in Dortmund.

Yours sincerely

Prof. Dr. Bernd Reusch


Committees
----------

Honorary Chairmen
D. B. Fogel, USA
T. Kohonen, Finland
Z. Pawlak, Poland
L. A. Zadeh, USA

General Chairman
B. Reusch, Dortmund


Programme Committee

Chairmen
K. Goser, Dortmund
H. Kiendl, Dortmund
B. Reusch, Dortmund
H.-P. Schwefel, Dortmund

Member
Aguilar-Martin, J.; France
Aizenberg, I.; Ukrain
Albrecht, R.; Austria
Bäck, Th.; Germany
Baldwin, J.; UK
Banzhaf, W.; Germany
Berenji, H. R.; USA
Borisov, A. N.; Latvia *
Bosc, P.; France
Bouchon-Meunier, B.; France
Brauer, W.; Germany
Carlsson, C.; Finland
Chan, L.; Hong Kong *
Damiani, E.; Italy *
De Jong, K.; USA
Di Nola, A.; Italy
Drechsler, R.; Germany
Dubois, D.; France
Eklund, P.; Sweden *
Esteva, F.; Spain *
Fathi, M.; Germany
Fedrizzi, M.; Italy
Frank, P. M.; Germany
Freksa, Ch.; Germany
Fukuda, T.; Japan *
Glesner, M.; Germany
Gottwald, S.; Germany Grauel, A.; Germany
Gupta, M. M.; USA
Grzymala-Busse, J.; USA
Hajek, P.; Czech Republic
Halgamuge, S.; Australia*
Hata, Y.; JapanHirota, K.; Japan
Homaifar, A.; USA
Isermann, R.; Germany
Jamshidi, M.; USA *
Kacprzyk, J.; Poland *
Kasabov, N.; New Zealand *
Kaynak, O.; Turkey *
Kerre, E.; Belgium
Klement, E. P.; Austria
Koczy, L.; Hungary
Kruse, R.; Germany
Larsen, H. L.; Denmark
Liu, Y. M.; China *
Mamdani, E. H.; UK
Mansour, M.; Switzerland
Mesiar, R.; Slovak Republic
Mohammadian, M.; Australia
Moraga, C.; Germany
Nauck, D.; Germany
Negoita, M. Gh.; New Zealand
Novak, V.; Czech Republic
Oh, K.-W.; South Korea
Ollero, A.; Spain
Orlowska, E.; Poland
Pedrycz, W.; Canada *
Reuter, M.; Germany
Ruspini, E. H.; USA
Sipper, M.; Switzerland
Tanaka, H.; Japan
Trillas, E.; Spain
Valverde, L.; Spain
Varachiu, N.; Romania *
Verdegay, J. L.; Spain
Yager, R. R.; USA
Yamakawa, T.; Japan

* regional co-ordinator

Industry Track Committee
Bastian, A.; VW AG
Georg, E.; Endress + Hauser GmbH & Co.
Grollmann, J.; Siemens AG
Schäfer, R.; Roche Diagnostics GmbH
Schröder, H.-J.; VDO Adolf Schindling AG
Schwärtzel, H. J.; FAST
Schubert, A.; Dow Deutschland Inc.
Splettstößer, W.; Siemens AG
Tresp, V.; Siemens AG
Wagner, R.; Thomson Multimedia GmbH

Sponsoring Organisations
------------------------

Arbeitsgemeinschaft Fuzzy Logik und Softcomputing Norddeutschland (AFN)
Asian Pacific Neural Network Assembly (APNNA)
Berkeley Initiative in Softcomputing (BISC)
City of Dortmund
Dachverband Medizinische Technik (DVMT)
Deutscher Verband für Schweißtechnik e. V. (DVS)
Deutsche Gesellschaft für Logistik e. V. (DGfL)
Dortmund Chamber of Industry and Commerce (IHK zu Dortmund)
European Society for Fuzzy Logic and Technologies (EUSFLAT)
Federation of International Robot-Soccer Association (FIRA)
Forschungsgemeinschaft Bekleidungsindustrie e. V.
Forschungskuratorium Gesamttextil
International Fuzzy Systems Association (IFSA)
Neural Network Council Member Societies (IEEE-NNC)

Organising Institutions
-----------------------

University of Dortmund, Computer Science I
VDE/VDI-Society of Microelectronics, Micro- and Precision
Engineering (GMM)
Information Technology Society within VDE (ITG)
VDI/VDE-Society on Measurement and Control (GMA)

in co-operation with

German Informatics Society (GI)


Organising Committee
--------------------

Dr. N. Jesse, University of Dortmund, Computer Science I
Dr.-Ing. V. Schanz, ITG - Information Technology Society within VDE
Dipl.-Ing. R. Theobald, GMM - VDE/VDI-Society of Microelectronics,
Micro- and Precision Engineering


Conference Tutorials
--------------------

On Thursday, May 27, and Friday, May 28, a series of tutorials will take place
with emphasis on both "hot topics for the expert" and education for
practitioners.

No. Name of Course/ Instructor Time

Thursday, May 27

A1
Prozessoptimierung mit Neuronalen Netzen
15.00-18.30
Thomas Froese, ATLAN-tec KG, Willich
A2
Anwendung Künstlicher Intelligenz in Banken und
Versicherungen
15.00-18.30
Gerd Doeben-Henisch, Inst. für Neue Medien,
Frankfurt/M.
Joachim Hasebrook, Bankakademie e. V., Frankfurt/M.
Bernd Reusch, Universität Dortmund
A3
Bildverarbeitung in der Medizin und neue Methoden
der Informatik (in Planung)
15.00-18.30
Klaus Mathias und Mitarbeiter, Städt. Kliniken Dortmund
Bernd Reusch und Mitarbeiter, Universität Dortmund
B1
Granular Computing
15.00-18.00
Witold Pedrycz, University of Manitoba, Canada

Friday, May 28

A4
Entscheidungsunterstützung mit Computational
Intelligence
9.00-12.30
Bernd Reusch und Mitarbeiter, University of Dortmund
in Kooperation mit der IHK zu Dortmund)
B2
Fuzziness in Database Management Systems
9.00-12.00
Janusz Kacprzyk, Polish Academy of Sciences, Poland
B3
Probabilistic Uncertain Reasoning
9.00-12.00
Jeffrey Paris, University of Manchester, United Kingdom
B4
A Wavelet Neuron Guaranteeing the Global Minimum
and its Applications to the Prediction of
the System Behavior and the Image Processing
9.00-12.00
Takeshi Yamakawa, Kyushu Inst. of Technology, Japan
B5
Learning, Adaption and Evolution for the Intelligent
Robotic System
9.00-12.00
Toshio Fukuda, Nagoya University, Japan
B6
Data Mining with Computational Intelligence Methods
14.00-17.00
Rudolf Kruse, Detlef Nauck, TU Braunschweig, Germany
B7
Foundations and Effective Methodology
of Fuzzy Logic Application
14.00-17.00
Vilem Novak, Irina Perfilieva,
University of Ostrava, Czech Republic
B8
Current Trends and Soft Computing Techniques
in the Field of Medical Imaging
14.00-17.00
Yutaka Hata, Himeji Institute of Technology, Japan
B9
Research and Application Projects in Space
Applications
14.00-17.00
Hamid R. Berenji, NASA Ames Research Center, Canada
B10
An Introduction to Evolutionary Algorithms
14.00-17.00
Hans-Georg Beyer, University of Dortmund, Germany



Tutorial Descriptions

A 1 Prozeßoptimierung mit Neuronalen Netzen
Thomas Froese

Prozeßmodellierung, Qualitätssicherung bei Produkten und ökonomische
Prozeßoptimierung mit Neuronalen Netzen sind Schwerpunkte dieses
Tutoriums. Neben einer allgemeinen Einführung behandelt es die
Versuchsplanung, die Berücksichtigung von Zeitverhalten und den Aufbau von
kompletten Automatisierungskonzepten mit KNNís.
Zielgruppe des Seminars sind inbesondere MSR-Betriebsingenieure und
Verfahrenstechniker in F&E-Abteilungen. Folgende Aspekte werden behandelt:
- Einführung in die Grundlagen der Künstlichen Neuronalen Netze (KNN)
- Genetische Algorithmen
- Projektierung mit standardisierter Industriesoftware
- Probleme bei der Arbeit mit "Real-World"-Daten aus Prozessen
- Statistische Voruntersuchungen an Daten für die Modellierung
- Methoden der Plausibilitätsanalyse bei Prozeßdaten
- Statistische Validierung und Dokumentation von Neuronalen Modellen
- Typische Problemlösungen mit KNN in: Trocknern, Destillationskolonnen,
Ethanolanlagen, Crackern, Polymerreaktoren, Extrudern, Mischern und
komplexen mehrstufigen Prozessen
- Versuchsplanung mit speziellen KNNís
- Standardvorgehensweise zur Projektierung mit KNN-Modellen
- Modellgestützte Regelung mit Neuro-Modellen
- Parameteroptimierung an Neuro-Modellen von Prozessen
- Standardisierung von Projektabläufen mit speziellen Softwarepaketen.

A 2 Anwendung Künstlicher Intelligenz in Banken und Versicherungen
Gerd Doeben-Henisch, Joachim Hasebrook, Bernd Reusch

"Künstliche Intelligenz", das war lange nur ein Thema für
Science-Fiction-Romane und akademische Forschungslabors; doch nun
verbreitet sie sich in der Industrie und in den Research-Abteilungen der
Banken und Versicherungen. Zusammen mit Intranets und weltumspannenden
Informationsnetzwerken steht die Künstliche Intelligenz (hier: Computational
Intellegence) nun - nach Multimedia - an der Schwelle zur Breitentechnologie:
Ob Fuzzy Logik in Camcordern oder Intelligente Agenten für Softwarepakete:
CI begegnet uns in immer mehr Bereichen.

Der Workshop ist speziell auf Entscheidungsträger in Banken und
Versicherungen zugeschnitten. Er ist aber auch geeignet für Personen, die an
Computational Intelligence Interessiert sind. Die Teilnehmer erhalten einen
Einblick in die CI-Grundlagen, einen Überblick über CI-Anwendungen in
Banken und Versicherungen sowie einen Ausblick auf aktuelle und zukünftige
Entwicklungen (lernende Strukturen). Alle Programmpunkte werden durch
Demonstrationen am Rechner begleitet.

A 3 Bildverarbeitung in der Medizin und neue Methoden der Informatik
Klaus Mathias und Mitarbeiter, Bernd Reusch und Mitarbeiter

In der medizinischen Diagnose spielen bildgebende Verfahren wie die
Computertomographie (CT), die Kernspintomographie
(Magnetresonanztomographie, MRT) und die Ultraschallanalyse eine
zunehmend bedeutsame Rolle. Die Auswertung dieser Bilder ist allerdings nicht
selten zeitaufwendig und mit beträchtlichen Unwägbarkeiten verbunden. Von
einer rechnergestützten Bildanalyse wird erwartet, daß sie schnellere und
objektivere (i. S. v. reproduzierbare) Ergebnisse liefert, als dies menschliche
Experten vermögen. Dieses Ziel kann nur erreicht werden, wenn Unsicherheiten
und Unschärfen in den Bildern vom Computer erkannt und angemessen
behandelt werden. Der Wunsch nach leistungsfähigen Verfahren zur
automatischen Objekterkennung steht aber auch mit der wachsenden Menge
an Bildmaterial und dem geplanten Aufbau multimedialer medizinischer
Informationssysteme in Zusammenhang. In diesem Tutorium wird aufgezeigt,
daß Methoden der sog. Computational Intelligence wesentliche Fortschritte in
der medizinischen Bildanalyse versprechen.

B 1 Granular Computing
Witold Pedycz

The objective of this tutorial is to provide a solid and comprehensive
introduction to a new paradigm of granular computing. Granular computing
emerges as a general way of computation and interpretation aimed at
processing granular information. The underlying concept permeating the entire
area is the one of information granules and information granulation. We
discuss the most commonly encountered vehicles of information granulation
such as set theory (especially interval analysis), fuzzy sets, rough sets,
shadowed sets, random sets, etc. The tutorial elaborates on these formal
platforms in more detail and contrast them with respect to the key properties of
the resulting information granules (including their semantics and robustness
aspects). Furthermore, we discuss the underlying processing principles that are
unique to the assumed formalism of information granulation. An intriguing
question dealing with a way of translation of information granules between
processing frameworks exhibiting different levels of information granularity
will be discussed in detail. This comes hand in hand with the problem of
representing any new data (no matter whether numeric or granular) in terms of
some generic information granules that are provided in advance.

Owing to its inherent nature, granular computing promotes a notion of
relational rather than function-driven mappings. This contributes to the
increased generality of the resulting constructs that are often geared towards
revealing associations between information granules rather than confining to
standard linear or non-linear regression-like dependencies. Finally, the tutorial
will concentrate on granular neuro-computing and granular rule-based
architectures as the two commonly encountered examples of granular
computing.

A 4 Mathematische Methoden zur Entscheidungsunterstützung
Bernd Reusch und Mitarbeiter
(in Kooperation mit der IHK zu Dortmund)

Verfahren zur Entscheidungsunterstützung, wie z.B. Simulationsverfahren, die
Methoden des Operations Research und der multi-kriteriellen
Entscheidungsfindung - um nur einige zu nennen -, besitzen einen überaus
breiten Einsatzbereich. Die Anwendungsgebiete reichen von der Unterstützung
strategischer Entscheidungen im Management über die Planung von Anlagen
und Produktionsprozessen bis zum Bereich der Steuer- und Regelungstechnik.
Konventionelle Methoden sind häufig auf exakte mathematische Modelle
angewiesen, die in der betrieblichen Praxis nicht oder nur mit großem Aufwand
erstellt werden können. Computational Intelligence, insbesondere die Fuzzy
Logik, setzt hier an: sie hilft, ungenaue oder unvollständige Informationen
mathematisch präzise zu verarbeiten. In Verbindung mit Neuronalen Netzen
oder Evolutionären Algorithmen kann Modellwissen aus vorhandenen
Datensätzen auch automatisch generiert werden.
Im Bereich strategischer Entscheidungen lassen sich ökonomische
Standardmodelle um scharfe oder unscharfe Komponenten der Wissens
erweitern und damit leichter an aktuelle Entwicklungen anpassen. Die
Bewertung aktueller oder zukünftiger Entwicklungen ist zentrale Grundlage für
Entscheidungen: hier können Methoden der multi-kriteriellen
Entscheidungsfindung oder auch Evolutionäre Algorithmen den
Entscheidungsträger unterstützen.
Bei der Planung von Produktionsstandorten und -abläufen können schon in
einer frühen Phase durch Fuzzy-Logik-basierte Simulation von
Betriebszuständen oder Nutzerverhalten Probleme erkannt und leistungsfähige
Lösungen entwickelt werden.
Im Bereich der Prozeßüberwachung und -steuerung werden bereits seit einigen
Jahren erfolgreich Microcontroller eingesetzt, die auf Fuzzy Logik beruhen; in
zunehmendem Maße werden auch Neuronale Netze berücksichtigt. Diese
Systeme finden dann Verwendung beispielsweise in der automatischen
Qualitätskontrolle, z. B. bei der Qualitätsüberwachung mittels Bildverarbeitung
und -bewertung.
Das Tutorium vermittelt einen grundlegenden Einblick in die Methoden der
Computational Intelligience und ihre praktischen Einsatzgebiete zur
Entscheidungsunterstützung.

B 2 Fuzziness in Database Management Systems
Janusz Kacprzyk

Past trends have greatly enhanced the usersí demand for more human
consistent and easy to use DBMSs. A crucial issue in this respect is the ability to
store and handle imprecise data resulting, e. g. from the use of natural
language.
This tutorial will start with a brief summary of basic approaches to DBMS,
concentrating on the relational model. Various aspects of data imperfection,
including uncertainty, imprecision (vagueness), ambiguity, etc., concentrating
on imprecision, will be considered.
Two basic lines of reasoning in the fuzzification of DBMSs will be introduced:
(1) the development of a "fuzzy DBMS" with tools for the representation of
fuzzy data and querying with fuzzy requests, and (2) the development of
"add-ons" to conventional (non-fuzzy) databases, mainly for querying. Fuzzy
querying will be shown to provide a tool for the implementation of intelligent
linguistic summaries of (large data sets in) databases. This will be illustrated on
an example of implementation of such a system for the summarization of sales
data at a computer retailer. The lecturer will discuss various solutions to the
fuzzification of querying and query languages and show some practical
solutions of fuzzy querying for some popular DBMSs, notably Microsoft Access
and FoxPro for Windows.
Future perspectives for the use of fuzzy logic in DBMS will be outlined,
advocating that this may be a future popular fuzzy product on the market.

B 3 Probabilistic Uncertain Reasoning
Jeffrey Paris

Currently the study of uncertain reasoning seems to be separating into two
main areas or disciplines. The first, and by far the most traditional, is
probabilistic uncertain reasoning, which is largely characterised as the study of
belief as subjective probability. The second, and much more recent, is called
Fuzzy Logic, which has close connections with multi-valued logics and
typically concerns itself with degrees of truth.
In contrast to the main thrust of this conference this tutorial will focus on the
first of these, probabilistic uncertain reasoning. In particular foundational
issues will be treated; why beliefs should change them in the light of new
information. So this tutorial will not tell you how to build an expert system, but
might better equip you to criticise (and appreciate) the efforts of others!
Prerequisites are familiarity with basic propositional calculus and feeling
comfortable expressing ideas within the language of mathematics.

B 4 A Wavelet Neuron Guaranteeing the Global Minimum and its
Applications to the Prediction of the System Behavior and the Image
Processing
Takeshi Yamakawa

A novel neuron model is presented which has non-linear synapses and
guarantees global minimum. Each non-linear synapse is represented by an
over-complete number of wavelets, the basis function of which is a constant in
level 0 and cosine functions in higher levels. Each basis function of cosine is
defined with a compact support. The wavelet neuron possesses distinctive
features of
(1) guarantee of the global minimum (i.e. optimally designable),
(2) much shorter learning time (typically three decades shorter) and much less
error than the multi-layer neural networks,
(3) more smooth than the traditional wavelets, etc.
The wavelet neuron is applied to the prediction of the behavior of a chaotic
system and image processing of noise filtering and edge enhancement.

B 5 Learning, Adaption and Evolution for the Intelligent Robotic System
Toshio Fukuda

This tutorial will provide the background of the learning, adaption and
evolution methods using the Fuzzy Logic, Neural Network and Evolutionary
Computation method for the intelligent robotic systems. In particular, the
hybrid system obtained by those methods will be explained clearly, including
the advanced architecture of the system, so that the robotic system can be
made more intelligent in comparison with the conventional methods. This
methodology can be applied to many industrial sectors.
The robotic system includes the fields of the motion control, trajectory control,
behavior control, hierarchical system control, teleoperation, distributed robotic
system and human-robot interface. This tutorial will show how the
performance can be improved by employing these proposed methods. Some of
the examples will be shown in the videos.

B 6 Data Mining with Computational Intelligence Methods
Rudolf Kruse, Detlef Nauck

In many industrial and commercial areas large amounts of data are gathered.
For many tasks like forecasting, marketing, quality control, etc. it is crucial to
convert data into knowledge. However, standard data analysis techniques like
simple statistical approaches often do not provide sufficient information to
support decision making.
Knowledge discovery in databases (KDD) means to identify valid, useful,
meaningful, unknown, and unexpected relationships in data bases. The term
data mining is used to describe the application of different learning methods
and analysis techniques in order to search for knowledge in data. This means,
data mining can be seen as a tool in a KDD process to automatically obtain
prognostic information from large data collections.
Data mining cannot be understood as a single method. There are many
different techniques that are used in data mining, for example, statistics,
machine learning, probabilistic networks, neural networks, fuzzy systems and
combinations of these, like neuro-fuzzy systems.
This tutorial introduces several computational intelligent data mining
techniques and provides some guidelines on how to select appropriate data
analysis methods depending on criteria like available training data, precision
and readability of the expected solution, usability and flexibility. By using
industrial examples from management and engineering, the possibilities and
problems of data mining will be illustrated.

B 7 Foundations and Effective Methodology of Fuzzy Logic Applications
Vilem Novak and Irina Perfilieva

The tutorial is devoted to the explanation and overview of some of the
up-to-date achievements in the theory, and to the proposal of the rational
methodology based on them. The latter will lead to effective applications of
fuzzy logic in various areas. Some methodological results will be demonstrated
on simulations.

The main topics covered in the tutorial are the following:

Methodology:
- A unifying view on different models for approximate representation of
functions: wavelets, neural networks and fuzzy logic rule bases of three basic
types, namely linguistic, singleton and Takagi-Sugeno ones,
- normal forms of fuzzy logic rule bases; what can be represented by them,
- methodology of approximate representation of continuous functions by
fuzzy logic normal forms; complexity and smoothness of the representing
models,
- effective and rational methodology for precise representation of continuous
functions,
- comparison of different models for applications to control.

Theoretical foundations:
- what is fuzzy logic, what can it bring to the practice,
- evaluating linguistic expressions and their elaboration,
- the structure of truth values with basic operations (minimum, maximum and
negation are not the only possibility); what are the reasons for such choice,
- what is logical deduction; two basic kinds of inference in linguistic fuzzy
logic rule base: logical deduction and fuzzy approximation; demonstration of
the behaviour and some methodological conclusions.

B 8 Current Trends and Soft Computing Techniques in the Field of
Medical Imaging
Yutaka Hata

This tutorial will provide an introduction to medical imaging and highlight the
soft computing techniques. The tutorial will start with a brief introduction
emphasizing that medical imaging techniques have received considerable
attention to enhance the quality of clinical diagnosis as the numbers of CT, MRI
and SPECT devices become popular. The reason why fuzzy medical imaging
techniques is useful to this area will be explained. Specific methods to be
reviewed include fuzzy rule-based segmentation procedure of MRI (magnetic
lever, neural computing based segmentation procedure of MRA (magnetic
resonance angiography) and a registration system of human skull obtained
from CT (Computed tomography) and blood vessel obtained from MRA. These
methods are developed in the laboratory of the speaker and are now practically
used in clinical spot.

B 9 Research and Application Projects
Hamid R. Berenji

In the first section of this tutorial, our current research projects are described
including Reinforcement Learning and its extension to Fuzzy Reinforcement
Learning (FRL), Multi agent learning and problem solving, and biologically
inspired intelligent systems. In the second part current application projects at
our group at NASA on implementing some of the results of the above research
in space applications are presented. This section will include a detailed
description of our work on the Space Shuttle Training Aircraft (STA) project,
multi-agent learning and problem solving for exploration of Mars and Jupiter's
moon Europa, and the hardware development of our two-legged walker. The
last section of this tutorial will describe the potentials of using computational
intelligence in development of Space and commercial intelligent systems.

B 10 An Introduction to Evolutionary Algorithms
Hans-Georg Beyer

Evolutionary Algorithms (EA) - the unifying term for Genetic Algorithms (GA),
Evolution Strategies (ES), and Evolutionary Programming (EP) - have received
considerable attention during the last decade. Gleaned from biological
metaphors, they are intended to serve as general purpose easy-to-use
optimization techniques capable of reaching globally optimal or at least nearly
optimal solutions. This is realized by biologically inspired variation and
selection rules. These rules are applied to a population (or several
sub-populations) of candidate solutions (individuals) that are evaluated with
respect to their fitness. Thus, it is possible by an evolutionary loop to
approximate successively the optimal state of the system to be investigated.
Due to their robustness, EA are well suited techniques for industrial design and
management tasks. They do not need gradient information and they can
operate in each kind of parameter space (continuous, discrete, combinatorial,
or even mixed variants). By means of the population concept EA easily can be
parallelized. This is why they often exhibit superior speedup behavior
compared to traditional optimization techniques.
The tutorial gives an introduction into the mainstream EA classes. Emphasis
will be laid on the basic algorithms (ES, EP, GA) including a discussion of
different variants of the evolutionary operators mutation, recombination, and
selection. Some theoretical issues will be raised aiming at a deeper
understanding of why and how the operators and the algorithms as a whole do
work.


Conference
----------

Tuesday, May 25
8.00 - 9.20
Registration and Morning Coffee

Inaugural addresses

9.20 - 9.30
Prof. Dr. B. Reusch, General Chairman
9.30 - 9.40
Prof. Dr. A. Klein,
Rector of the University of Dortmund
9.40 - 9.50
Mr Karl Schultheis,
Ministry for Schools, Higher Education, Science and Research
of North-Rhine Westphalia
9.50 - 10.00
Dr. rer. pol. H. Feucht, IBM Deutschland GmbH,
for the Learned Societies
10.00 - 11.00
Invited Lecture 1
Spotting Relevant Information in Extremely Large
Document Collections
T. Kohonen, Helsinki University of Technology, FIN

Session 1A
Session Chair: A. N. Borisov

11.00 -11.30
Similarity Based System Reconfiguration by Fuzzy
Classification and Hierarchical Interpolate Fuzzy
Reasoning
S. Kovács, University of Miskolc, H
11.30 - 12.00
A Fuzzy System for Fetal Heart Rate Assessment
J. F. Skinner, M. Garibaldi, E. C. Ifeachor,
University of Plymouth, UK

Session 1B
Session Chair: G. Rudolf

11.00 - 11.30
Efficient Graph Coloring by Evolutionary Algorithms
N. Drechsler, W. Günther, R. Drechsler
Institute of Computer Science, Freiburg im Breisau, D
11.30 - 12.00
Determination of Decision Rules on the Basis of Genetic
Algorithms
A. Takahashi, A. N. Borisov,
Technical University of Riga, LV

Session 1C
Session Chair: C. Moraga

11.00 - 11.30
Modeling a Refrigeration System Using Recurrent Neural
Networks
R. Habtom, University of Kaiserslautern, D
11.30 - 12.00
Evaluating Nugget Sizes of Spot Welds by Using Artificial
Neural Network
T. Yiming, F. Ping, Z. Yong, Y. Siqian,
Northwestern Polytechnical University, Xiían, CN
12.00 - 13.00
Lunch (On Your Own)
13.00 - 14.00
Invited Lecture 2
Topological Theory of Fuzziness
R. Albrecht, University of Innsbruck, D

Session 2A
Session Chair: W. Pedrycz

14.00 - 14.30
Fuzzy Controller Generation with a Fuzzy Classification
Method
I. Borgulya, Janus Pannonius, University,Pécs, HU
14.30 - 15.00
Transformation and Optimization of Fuzzy Controllers
Using Signal Processing Techniques
F. Fernández Hdz.,J. Gutiérrez, Departamento de Tecnologia
Fotónica, Madrid, E
15.00 - 15.30
Fuzzy-Control Design Tool for Low-Cost Microcontrollers
(FHFC-Tool)
K.-D. Kramer, J. Kirschner, S. Wöhlbier,
Hochschule Harz,Wernigerode, D

Session 2B
Session Chair: H.-P. Schwefel

14.00 - 14.30
Search of Optimal Error Correcting Codes with Genetic
Algorithms
J. Lacan, P. Chatonnay, IUT Belfort-Montbéliard, F
14.30 - 15.00
An Unsupervised Clustering with Evolutionary Strategy to
Estimate the Cluster Number
K. Imai, N. Kamiura, Y. Hata,
Himeji Institute of Technology, J
15.00 - 15.30
Multi-Objective Optimization in Evolutionary Algorithms
Using Satisfiability Classes
N. Drechsler, R. Drechsler, B. Becker,
Institute of Computer Science, Freiburg im Breisgau, D

Session 2C
Session Chair: M. Fathi

14.00 - 14.30
Neural Network Approach to Design of Distributed Hard
Real-Time Systems
J. Martyna, Jagiellonian University, Krakow, PL
14.30 - 15.00
Supporting Traditional Controller of Combustion Engines
by Means of Neural Networks
C. Ungerer, TU München,
D. Stübener, Kratzer Automation, D
C. Kirchmair, M. Sturm, TU München, D
15.00 - 15.30
Controlling Biological Wastewater Treatment Plants Using
Fuzzy Control And Neural Networks
M. Bongards, FH Köln - Gummersbach, D
15.30 - 16.00
Coffee Break

Session 3A
Session Chair: Y. Hata

16.00 - 16.30
A New Fuzzy Character Segmentation Algorithm for
Persian / Arabic Typed Texts
M. B. Menhaj, F. Razzazi,
Amirkabir University of Technology, IR
16.30 - 17.00
RePART: A Modified Fuzzy ARTMAP for Pattern
Recognition
A. Canuto, G. Howells, M. Fairhurst,
University of Kent at Canterbury,UK
17.00 - 17.30
An Adaptive C-Average Fuzzy Control Filter for Image
Enhancement
F. Farbiz, M. B. Menhaj, S. A. Motamedi, Amirkabir
University of Technology, IR

Session 3B
Session Chair: M. G. Negoita

16.00 - 16.30
Pareto-optimality in Scheduling Problems
A. F. Gómez-Skarmeta, F. Jiménez, J. Ibáñez,
Universidad de Murcia, E
16.30 - 17.00
Controlled Markov Chain Optimization of Genetic
Algorithms
Y. Cao, University of the West of England, Bristol, UK
L. Cao, New Jersey Institute of Technology, Newark, USA
17.00 - 17.30
Tackling Epistatic Problems using Dynastically Optimal
Recombination
C. Cotta-Porras, J. M. Troya,
University of Málaga, E

Session 3C
Session Chair: N. Varachiu

16.00 - 16.30
Extended Methods for Classification of Remotely Sensed
Images Based on ARTMAP Neural Networks
N. KopËo, Boston University, USA
P. SinËák, Technical University of Kosice, SQ
H. Veregin, University of Minnesota, USA
16.30 - 17.00
Application of Artificial Neural Network in Control of
Vector Pulse-Width Modulation Inverter
P. Brand1tetter, M. Skotnica,
Techn. University of Ostrava, CZ
17.00 - 17.30
Modeling multiple microstructure transformations in
steels with a Boltzmann neural net
E.-D. Schmitter, FH Osnabrück, D
19.30
Get Together
Additionally, it is intended to organise an event at the
Dortmund Chamber of Industry and Commerce.

Wednesday, May 26

8.30 - 9.00
Morning Coffee
9.00 - 10.00
Invited Lecture 3
Evolutionary Computation: Were we are and where we're
headed
K. De Jong, George Mason University, USA

Session 4A
Session Chair: T. Fukuda

10.00 - 10.30
Fuzzy Controllers by Unconventional Technologies for
Tentacle Arms
M. Ivanescu, V. Stoian, University of Craiova, RO
10.30 - 11.00
Control of Robot Arm Approach by Fuzzy Pattern
Comparison Technique
M. Bonkovic, D. Stipanicev, M. Stula,
University of Split, HR
11.00 - 11.30
A Fuzzy Shapes Characterization for Robotics
N. Varachiu, National Institute of Microtechnology - IMT
Bucharest, RO
11.30 - 12.00
ART-based Automatic Generation of Membership Functions
for Fuzzy Controllers in Robotics
G. Attolico, A. Itta, G. Cicirelli, T. DíOrazio,
Consiglio Nazionale Delle Ricerche, Bari, I

Session 4B
Session Chair: E. Kerre

10.00 - 10.30
On Two Types of LM Fuzzy Lattices
G. Trajkovski, B. ªukiÊ, West Virginia University, Morgantown,
USA
10.30 - 11.00
Generated Connectives in Many Valued Logic
Radko Mesiar, Slovak Technical University Bratislava, SLO
11.00 - 11.30
Conjugacy Classes of Fuzzy Implications
M. BaczyÒski, J. Drewniak,
Silesian University, Katowice, PL
11.30 - 12.00
Characterization of Dienes Implication
M. BaczyÒski, Silesian University, Katowice, PL

Session 4C
Session Chair: K. Goser

10.00 - 10.30
Neural Networks Based on Multi-Valued and Universal
Binary Neurons: Theory, Application to Image Processing
and Recognition.
I. N. Aizenberg, State Scientific and Research Institute of
Informational Infrastructure, Uzhgorod, UKR
10.30 - 11.00
Modeling of Thermal Two Dimensional Free Turbulent Jet
by a Three Layer Two Time Scales Constant Cellular
Neural Network
A. Shabani, M. B. Menhaj, H. B. Tabrizi, Amir Kabir University
of Technology, Tehran, IR
11.00 - 11.30
A Neural Approach for Detection of Road Direction in
Autonomous Navigation
V. Neagoe, M. Valcu, B. Sabac, Polytechnic University of
Bucharest, RO
11.30 - 12.00
A Neural Segmentation of Multispectral Satellite Images
V. Neagoe, I. Fratila,
Polytechnic University of Bucharest, RO
12.00 - 13.00
Lunch (On Your Own)

Session 5A
Session Chair: P. Bosc

13.00 - 13.30
Design of a Traffic Junction Controller Using Classifier
System and Fuzzy Logic
Y. J. Cao, N. Ireson, L. Bull, R. Miles, University of the West of
England, UK
13.30 - 14.00
Using Fuzzy Logic to Control Traffic Signals at
Multi-Phase Intersections
J. Niittymäki, Helsinki University of Technology, FIN
14.00 - 14.30
Fuzzy Control to Non-Minimal Phase Processes
W. Gharieb, Ain Shams University, Cairo, EG
14.30 - 15.00
Robust Un-Coupled Fuzzy Controller for Longitudinal and
Lateral Control of an AGV
K. R. S. Kodagoda, W. S. Wijesoma, E. K. Teoh, Nanyang
Technological University, Singapore, SGP

Session 5B
Session Chair: H. Kiendl

13.00 - 13.30
Center Manifold Theory Approach to the Stability and
Analysis of Fuzzy Control Systems
R.-E. Precup, S. Preitl, S. Solyom, Politechnic University of
Timisoara, RO
13.30 - 14.00
Stability Analysis of Fuzzy and Other Nonlinear Systems
based on the Method of Convex Decomposition
R. Knicker, University of Dortmund, D
14.00 - 14.30
Fuzzy utilities comparison in multicriteria analysis
H. Deng, Ch.-H. Yeh, Monash University, AUS
14.30 - 15.00
A Possibilistic Formalization of Case-Based Reasoning
and Decision Making
E. Hüllermeier, IRIT - Institut de Recherche en Informatique
de Toulouse, F

Session 5C
Session Chair: N. Kasabov

13.00 -
13.30
Parameter Determination for Nano-Scale Modelling
M. Thomas, C. Pacha, K. Goser, University of Dortmund, D
13.30 -
14.00
Optimizing Routing Algorithms in Telecommunication
Networks with Neural Networks and Fuzzy Logic
I. Gilsdorf, Siemens Business Services ,D
W. Brauer, TU München, D
14.00 -
14.30
Optimum Work Roll Profile Selection in the Hot Rolling of
Wide Steel Strip Using Computational Intelligence
L. Nolle, A. Armstrong, A. Hopgood,
The Open University in Wales, Cardiff, UKbr> A. Ware,
University of Glamorgan, Pontypridd, UK
14.30 -
15.00
Analyzing Epileptic Events On-Line by
Soft-Computing-Systems
M. Reuter, TU Clausthal/University of Dortmund, D
15.00 -
15.30
Coffee Break

Session 6A
Session Chair: L. C. Jain

15.30 - 16.00
On Interactive Linguistic Summarization of Databases via
a Fuzzy-Logic-Based Querying Add-On to Microsoft
Access
J. Kacprzyk, S. Zadrozny, Polish Academy of Sciences,
Warsaw, PL
16.00 - 16.30
A General Purpose Fuzzy Engine For Crop Control
M. Ahmed, King Fahd University of Petroleum and Minerals,
SA
E. Damiani, A. G. B. Tettamanzi, Università di Milano, I
16.30 - 17.00
Fuzzy Control of a Physical Double Inverted Pendulum
Model
J. Vascák, Technical University of Kosice, SK
17.00 - 17.30
Synthesis of Stable Fuzzy PD/PID control laws for Robotic
Manipulators from a Variable Structure Systems
Standpoint
W. S. Wijesoma, K. R. S. Kodagoda, Nanyang Technological
University, Singapore, SGP

Session 6B
Session Chair: I. Aizenberg

15.30 - 16.00
On data summaries based on gradual rules
P. Bosc, O. Pivert, IRISA/ENSSAT, Lannion, F
L. Ughetto, IRISA/IUT de Lannion, F
16.00 - 16.30
Working Towards Connectionist Modeling of Term
Formation
P. Marshall, Z. Bandar, Manchester Metropolitan University,
UK
16.30 - 17.00
About the Quantization of the Neural Nets
M. Reuter, TU Clausthal/University of Dortmund, D
17.00 - 17.30
Risk Analysis using Perceptrons and Quadratic
Programming
B.-J. Falkowski, FH Stralsund, D

Session 6C
Session Chair: L. Strobel Stewart

15.30 - 16.00
Finding Relevant Process Characteristics with a Method
for Data-based Complexity Reduction
j. Praczyk, H. Kiendl, T. Slawinski, University of Dortmund, D
16.00 - 16.30
Traffic Control in an ATM Network Using Rough Set
Theory
J. Martyna, Jagiellonian University, Krakow, PL
16.30 - 17.00
Evaluation og Characteristic Temperatures of Materials
Using Approximate Reasoning Method
B. Butkiewicz, T. Mroczek, Warsaw University of Technology,
PL
M. W. Grzybek, Industrial Institute of Electronics, Warsaw, PL
17.00 - 17.30
A Human Centered Architecture for Distributed Retrieval
of Medical Images
R. Castelletti, Università di Verona, I
E. Damiani, G. Righini, Universitá di Milano, I
R. Khosla, LaTrobe University, AUS
19.30
Conference Banquet

Words of Greeting
Lord Mayor Günter Samtlebe, City of Dortmund
President Dr. Winfried Materna, Dortmund Chamber of
Industry and Commerce

Evening Speech
From the Hanseatic League to an Industrial Metropolis
apl. Prof. Dr. phil. G. Sollbach, University of Dortmund
Location: City Hall

Thursday, May 27

8.30 - 9.00
Morning Coffee
9.00 - 10.00
Invited Lecture 4
A New Direction in Fuzzy Logic - Toward a Computational
Theory of Perceptions
L. A. Zadeh, University of Berkeley, USA

Session 7A: Industry Session
Session Chair: B. Reusch

10.00 - 10.30
Rezepturoptimierung mit Künstlichen Neuronalen Netzen
Th. Froese, Atlan-tec KG, Willich, D
10.30 - 11.00
Neuronale Netze als Alternative zur Versuchsplanung in
der Verfahrensentwicklung und dem Anfahren von
Neuanlagen
Th. Froese, Atlan-tec KG, Willich, D
11.00 - 11.30
Fuzzygestütztes Bewertungsmodell für die Entwicklung
umweltverträglicher Produkte und Prozesse
A. Schwan, Univ. of Erlangen-Nürnberg./ VW AG
11.30 - 12.00
Professional Judgement in Risk Assessment
L. Strobel Stewart, University of Manitoba, CA

Session 7B
Session Chair: H. R. Berenji

10.00 -
10.30
Representing the Real-Time Behaviour of Technical
Processes in Neural Nets by Using the Phase-Space-Flow
of the Degrees of Freedom of the Nets
M. Reuter, D. P. F. Möller, University of Hamburg, D
10.30 -
11.00
Spatial neural networks based on fractal algorithms
Biomorph nets of nets of .....
Th. Kromer, Münsterklinik Zwiefalten, D
11.00 -
11.30
Adaptive Control Systems Based on Neural Networks
L. Dafinca, Transilvania University, Brasov, RO
11.30 -
12.00
Function Approximation using Tensor Product Bernstein
Polynomials - Neuro & Evolutionary Approaches
M. Buzoianu, F. Oltean, A. Agapie,
National Institute of Microtechnology-Bucharest, RO

Session 7C
Session Chair: T. Yamakawa

10.00 - 10.30
Tuning Considerations above a Fuzzy Controller Used for
the Synchronous Generator
I. Filip, O. Prostean, D. Curiac,
Politechnic University of Timisoara, RO
10.30 - 11.00
A Sectoring Genetic Algorithm for the Urban Waste
Collection Problem
M. T. Lamata, J. I. Peláez,Universidad de Granada, E
J. C. Sierra, J. M. Bravo, Universidad de Málaga, E
11.00 - 11.30
Parameter Optimization of Group Contribution Methods
in High Dimensional Solution Spaces
C. Kracht, H. Geyer, P. Ulbig, S. Schulz,
University of Dortmund, D
11.30 - 12.00
Applying Heuristic Algorithms on Structuring
Europeanwide Distribution Networks
R. Djamschidi, A. Bruckner, RWTH Aachen, D
12.00 - 12.30
Closing Session with Best Paper Award


Poster Presentation
-------------------

In parallel with the talks posters will be presented, enabling researchers to
discuss late-breaking results, significant work in progress, or work that is best
communicated through conversation. Poster presentations allow conference
participants to exchange ideas one-on-one with authors, and let authors
discuss their work in detail with those participants most deeply interested in
the same topic.

Among other things, the following posters will be presented:

The Control of a Nonlinear System Using Neural Networks
Ovidi Grigore, Octavian Grigore, Politechnic University of Bucharest, RO

Demonstration of the Use of Hyperinference: Position Control Using an
Electric Disc-Motor
P. Krause, University of Dortmund, D

Synthesis of knowledges based on inductive generalization
S. Astanin, T. G. Kalashnikova, Taganrog State University of Radio Engineering,
RU

ANN-Based Fault Type and Location Identification System for
Autonomous Preventive-Restoration Control of Complex Electrical Power
System Plants
A. Halinka, M. Szewczyk, B. Witek, Silesian Techn. University, Gliwice, PL

Optimal Hidden Structure for Feedforward Neural Networks
P. Bachiller, R. M. Pérez, P. MartÌnez, P. L. Aguilar, P. DÌaz, University of
Extremadura, E

Relation Between the IADL and Physical Fitness Tests Focussing on
Uncertainty of Answering Questionnaire in Elderly Women
H. Uchida, Y. Hata, K. Suei, H. Nakagawa, Himeji Institute of Technology,
J.H. Aoyama, Okayama University Medical School, J

A Toolbox of Artificial Brain Cells to Simulate Classical and Operand
Learning Behavior
G. Döben-Henisch, Institut für Neue Medien, Frankfurt,
DJ. Hasebrook, Bankakademie e. V., Frankfurt, D

Intelligent, ANN-Based Identification System of Complex Generating Sets
Operating Mode
A. Halinka, M. Szewczyk, B. Witek, Silesian Techn. University, Gliwice, PL

A neural network based on linguistic modifiers
M. De Cock, F. Vynckier, E. Kerre, University of Gent, B

Fuzzy Computing in a MultiPurpose Neural Network Implementation
C.-D. Neagu, V. Palade, University "Dunarea de Jos" of Galati, RO

Fuzzy Logic Control of Child Blood Pressure During Anesthesia
A. Yardimci, A. S. Onural, Akdeniz University, TR

Soft Computing Genetic Tool V 3.0-Applications
A. H. Dediu, A. Agapie, N. Varachiu, National Institute of Microtechnology,
Bucharest, RO

Learning Experiments with CMOS Artificial Neuron
V. Varshavsky, V. Marakhovsky, The University of Aizu, J

Evolving Fuzzy Connectionist Systems for On-line Learning: Methods,
Tools, Applications
N. Kasabov, University of Otago, NZ

Automated fluorescent scanning microscopy: Cell classification by fuzzy
functions
S.V. Varga, B. Molnar, G. Csendes, R. Schaefer* , W. Mahoney**,
Cell Analysis Laboratory, II.Dept. of Medicine,
Semmelweis Uni, Budapest, Hungary
*Roche Diagnostics Tutzing,Germany, **Berkeley, Ca. USA

If you would like to present a poster, please send a short abstract to the
conference secretariat.

Exhibition
----------

The conference incorporates an exhibition featuring hardware, software and
applications. This exhibition offers an excellent opportunity to present your
developments to a highly interested audience. If you are interested in
participating in the exhibition, please contact the Conference Secretariat for
further information.

2nd FIRA Robot Soccer Eurocup 1999
----------------------------------

In conjunction with 6th Fuzzy Days the 2nd FIRA Robot Soccer European Cup
will be held in Dortmund.

Robot Soccer Competition gives an opportunity to foster intelligent techniques
and intelligent robotic research by providing a standard problem where a wide
spectrum of technologies such as collaborative multiple agent robotics,
autonomous computing, real-time reasoning and sensor fusion can be
developed, tested and integrated. Games under the auspices of FIRA shall
promote developments in autonomous soccer robots and intelligent systems
that can co-operate with each other.

http://ls1-www.cs.uni-dortmund.de/fira99

General Information
-------------------

Homepage of the Conference

http://ls1-www.cs.uni-dortmund.de/fd6

Conference Location

The Conference, the Exhibition and Poster Presentation will take place in the

Audimax, University Campus,
Vogelpothsweg 87,
D-44227 Dortmund.

The Tutorials will be held at the

Steigenberger Maxx Hotel,
Berswordtstraße 2,
D-44139 Dortmund,
Phone +49 231 9021 0
Fax +49 231 9021 999.

Conference Secretariat
All correspondence and requests for information should be addressed to

Mrs Ulrike Lippe
University of Dortmund
Computer Science I
D-44221 Dortmund/Germany
Phone : +49 231 755 6223
Fax : +49 231 755 6555
E-mail : fd6@ls1.cs.uni-dortmund.de

Registration Desk

Registration Desk during the Conference

in the Audimax, University Campus (for the conference)
Tue - Wed: 8:00 - 18:00
Phone +49 231 755 6055,
Fax +49 231 755 6055

at the Steigenberger Maxx Hotel (for the tutorials)
Thursday, May 27, 14:00 - 18:00
Friday, May 28, 8:00 - 18:00
Phone +49 231 9021 500
Fax +49 231 9021 999

Registration Information and Conference Fees

To register please fill in the enclosed form and return by mail or fax in
conjunction with payment. If you have not paid prior to the conference, you
will be requested to settle the full amount at the registration desk.
A pre-registration is required for all tutorials prior to May 8, 1999.
Conference fees must accompany registration and be paid in full in Deutsche
Mark (DM).

prior to May 8 from May 8
University fees 585,-- 645,--
University fees:
GMM-, GI-, GMA- or ITG-member 525,-- 585,--
Full registration 895,-- 985,--
Industry fees:
GMM-, GI-, GMA- or ITG-member 815,-- 895,--
Student fees 150,-- 150,--
Tutorial fees (per tutorial) 480,--

Exhibitors must pay the registration fee for one person and a flat rate of DM
200,--; one additional person is free of charge.

Full, university and member registration includes admission to all sessions and
the exhibition area, morning and afternoon refreshments and a copy of the
conference proceedings. Also, the right to purchase additional conference
proceedings at a discount of 47,5 % off the list price. Member rates require
your membership number for one of the listed institutions.
Student registration includes admission to all sessions and the exhibition area,
morning and afternoon refreshments and the right to purchase the conference
proceedings at a discount of 47,5 % off the list price. (Studentsí fees do not
include a copy of the proceedings.) Student rates require a copy of your valid
full-time (!) student ID.

Tutorial registration includes lunch, refreshments and tutorial notes. The right
to purchase the conference proceedings at a discount of 47,5 % off the list price
is included.
Space for tutorials is limited and enrolment will be processed on a "first come,
first serve" basis.
The fee should preferably be paid by bank order (free of any bank transfer
charges) to account:

VDE - Verband der Elektrotechnik, Elektronik, Informationstechnik
Bethmann Bank, Frankfurt/M.
BLZ 501 301 00
Account-No. 46754-2-03
Keyword: Fuzzy

Credit card facilities are also available and the following cards will be accepted
for payment: American Express, VISA and MasterCard. Cheques will not be
accepted.
A receipt of payment will be sent to confirm registration.

Refund Policy

Written requests for cancellation must be postmarked no later than May 8,
1999. Refunds are subject to a DM 70,-- processing fee. No refund is possible
for cancellations received after this date. Refunds will be issued six to eight
weeks after the conference. Substitutions will be accepted at any stage.

Conference Proceedings

The conference proceedings is published by Springer Verlag in the series
Lecture Notes in Computer Science and will be distributed at the registration
desk.

Language

Conference language is English. Language for tutorials A1 - A 4 is German.

Accommodation

Hotel accommodation at preferential rates for participants have been arranged
at

Steigenberger Maxx Hotel***, Phone: +49 231 902 10, Fax:+49 231 902 1999,
Sol Inn Hotel***, Phone: +49 231 97 050, Fax:+49 231 970 5444,
Astron Suite Hotel****, Phone: +49 231 90 550, Fax:+49 231 905 5900
Hotel Atlanta * Phone: +49 231 55 70 75 0, Fax:+49 231 58 60 054

Please use Keyword "Fuzzy"!

Alternatively, you can contact:

Dortmunder Verkehrsverein e. V.
Königswall 20
D-44137 Dortmund
Phone +49 231 502 21 74
Fax: +49 231 16 35 93

############################################################################
This message was posted through the fuzzy mailing list.
(1) To subscribe to this mailing list, send a message body of
"SUB FUZZY-MAIL myFirstName mySurname" to listproc@dbai.tuwien.ac.at
(2) To unsubscribe from this mailing list, send a message body of
"UNSUB FUZZY-MAIL" or "UNSUB FUZZY-MAIL yoursubscription@email.address.com"
to listproc@dbai.tuwien.ac.at
(3) To reach the human who maintains the list, send mail to
fuzzy-owner@dbai.tuwien.ac.at
(4) WWW access and other information on Fuzzy Sets and Logic see
http://www.dbai.tuwien.ac.at/ftp/mlowner/fuzzy-mail.info
(5) WWW archive: http://www.dbai.tuwien.ac.at/marchives/fuzzy-mail/index.html