IFSA Newsletter Vol.2, No.3
Issued on Aug. 31, 2005
http://www.cmplx.cse.nagoya-u.ac.jp/~ifsa/

It is my genuine honor and great responsibility to serve as an IFSA President for the term 2005-2007. I feel very much privileged to work with the IFSA Council - a team of highly dedicated, knowledgeable, and determined to excel individuals. The current IFSA officers are

President Elect: Janusz Kacprzyk

Past President : Zenn Bien

Vice President, Awards: Kaoru Hirota

Vice President, Publicity: Oscar Castillo

VP President, Conferences and Technical Committees: Mingsheng Ying

Vice President, Membership: Imre Rudas

Secretary: Fernando Gomide

Treasurer: Kyung Chan Min

Our thanks go to the past Council, especially its president Zenn Bien. I am sure that we will be continuously benefiting from his advice and support.

The agenda for this term is quite concise yet pretty challenging. In a nutshell, there are three major groups of pursuits

• Investing in the generation of young of researchers. What we have started this year at the IFSA Congress in the form of the best student awards and the IFSA-Elsevier young researcher award, will be continued, expanded and refined
• Expanding the membership and acknowledging superb accomplishments of IFSA members. In the years to come we hope to add new institutional members. The IFSA fellows program was successful in the past and we will be building on these successes.
• New services to the fuzzy set community. Our website will be augmented by adding a variety of educational material and publications. We will also concentrate on the coverage of the history of fuzzy sets and bring the legacy of the area to the IFSA community

Time will show to what extent those targets are accomplished; I do hope that two years from now we will be having a lot of developments to report on.

The IFSA 2005 Congress held in Beijing is over (which was a truly remarkable event in many different ways). It really helped us reflect upon the past (40 years of fuzzy sets and 20 years of IFSA- both being a very much spectacular anniversaries) and engage in making projections on the future of fuzzy sets and their underlying technology.

The preparations for the IFSA 2007 to be held in Cancun, Mexico are already underway; please check for the recent updates at http://www.hafsamx.org/ifsa2007/. And, as always, visit our Website for the recent updates on the upcoming events, community activities and other useful IFSA information http://www.cmplx.cse.nagoya-u.ac.jp/~ifsa/

As usual, all your input is essential - I am always reached at pedrycz@ee.ualberta.ca.

Witold Pedrycz
August 5, 2005

top

IFSA2005 World Congress successfully held in Beijing

The 11^th International Fuzzy Systems Association World Congress (IFSA2005) took place on July 28-31, 2005, in Tsinghua University, Beijing, with the themes of fuzzy logic, soft computing, and computational intelligence. This Congress, held for the first time in Mainland China, was organized by Fuzzy Mathematics and Fuzzy Systems Association of China (The IFSA China Chapter), Tsinghua University, and Sichuan University. Lotfi A. Zadeh served as Honorary Chair of the Congress, Yingming Liu as General Chair, Guoqing Chen as Organizing Chair, and Mingsheng Ying as Program Chair.

The Congress has been deemed a great success from various respects. First, it brought together scholars and practitioners from academia and industry to present the latest development in the theories and applications of fuzzy logic, soft computing, and computational intelligence. It attracted more than 340 delegates from over 40 countries and from every region of the globe. The program featured keynote/plenary speeches by distinguished scholars such as Lotfi Zadeh, Didier Dubois, George Klir, Yingming Liu, and Witold Pedrycz, along with 30 parallel ordinary and special sessions.

The conference proceedings were published in three volumes by Tsinghua-Springer with the accepted papers selected from over 620 submissions by a standard peer-review process. The subjects of the papers and scientific sessions ranged from theory to applications, reflecting a wide spectrum of coverage of the areas of fuzzy logic, soft computing, and computational intelligence, including foundations and mathematics, control, pattern recognition/image processing, databases and data mining, decision analysis, optimization, and related soft computing techniques such as neural networks, rough sets, genetic algorithms, and so on.

Second, a best paper award and five outstanding student papers awards were given at the conference. These awards were sponsored by IFSA, Elsevier, and the IFSA2005 organizers to encourage quality research and young scholars. The awards winners were selected by committees composed of internationally renowned scholars. In addition, IFSA named seven new IFSA Fellows, Piero Bonissone (USA), Bernadette Bouchon-Meunier (France), Christer Carlsson (Finland), Yingming Liu (China), Witold Pedrycz (Canada), Tomohiro Takagi (Japan), and Takashi Yamakawa (Japan), to bestow recognition on people who have made outstanding contributions to the field of fuzzy sets and related disciplines. Notably, Professor Yingming Liu, academician of the Academy of Science of China, is the first IFSA Fellow from a developing country.

Third, IFSA2005 was held in the 40^th year of fuzzy set theory and in the year of IFSA's 20^th anniversary. As an event of memory and celebration, the Congress featured an exhibition showing memorabilia from past IFSA congresses with pictures, posters, and collected materials including conference proceedings of all IFSA congresses. Moreover, a ceremonial session celebrated the 20^th IFSA anniversary and the former IFSA presidents were honored for their outstanding contributions and service to the society.

Finally, along with the high quality of the papers, sessions, and activities, the great success of IFSA2005 was also guaranteed by high-quality organizational arrangements and logistic operations, as well as sponsorship from China's Mathematics Research Center (Ministry of Education of China), China's National Natural Science Foundation, and Elsevier.

top

Libor Behounek, Petr Cintula
"General logical formalism for fuzzy mathematics: methodology and apparatus"

This paper, presented and selected as the best paper at the 11th IFSA World Congress, Beijing 2005, describes the theoretical foundations of a new project of the Prague research group in fuzzy logic directed towards developing and applying a rich logical apparatus for the systematization and formalization of fuzzy mathematics. The first steps in this direction were enabled by recent results in mathematical fuzzy logic, especially by the emergence of higher-order fuzzy logics, proposed by the authors. The core of the project is a formulation of certain formalistic methodology (based on Petr Hajek's approach to fuzzy logic in narrow sense), the development of suitable mathematical and logical tools, and the application of both the methodology and tools in various fuzzy disciplines (fuzzy probability, fuzzy numbers, fuzzy relations etc.). Employing the axiomatic approach enables a systematic study of fuzzy structures by proof-theoretic and model-theoretic methods; the unified formalism simplifies establishing interconnections of particular fuzzy disciplines, and aims at providing formal foundations for a large part of fuzzy mathematics.

Mr. Libor Behounek is a graduate student researcher at the Institute of Computer Science of the Academy of Sciences of the Czech Republic and a PhD student at Charles University in Prague, where he received his master's degree in logic in 2002. He has published several papers on fuzzy logic and fuzzy sets and has regularly participated at international conferences since 2003. His research is aimed at the logical foundations of fuzzy set theory and formal fuzzy mathematics.

Dr. Petr Cintula is a junior researcher at the Institute of Computer Science of the Academy of Sciences of the Czech Republic. He received his PhD from the Czech Technical University, Prague, in 2005. He is interested in mathematical aspects of fuzzy logic, has published more than 15 papers in scientific journals, and regularly lectures at international conferences since 2001. His recent research is focused on formal aspects of fuzzy mathematics and exploiting general methods for fuzzy logical calculi.

top

Ramin Halavati, "A Novel Noise Immune Approach to Speech Recognition"

Mr. Ramin Halavati received his M.S. in Artificial Intelligence from Sharif University of Technology, Tehran, Iran in 2003, and is now studying for a Ph.D. in artificial intelligence in the Artificial Creatures Lab, Computer Engineering Department, Sharif University of Technology under supervisorship of Dr. Saeed Bagheri Shouraki. His research interests are generally in cognition, learning, and consciousness resulting in pattern recognition, behavior generation, and machine learning.
He has worked on evolution of complex behaviors in artificial life simulations in his M.S. thesis, creating a test bed for cognitive studies called Zamin and evolving creatures capable of communication and producing artificial emotions.
His current research is on pattern recognition by generating hierarchies of sub-patterns and super-patterns that can lead to faster or more accurate recognition of patterns or increment of pattern learning speed by gaining more experiences from previous tasks. This system is being bench marked under speech recognition and cursive hand writing recognition domains.
He is the author of about 30 papers by now and expects to receive his Ph.D. in 2007.

Iman Karimi, "A Fuzzy-Probabilistic Risk Assessment System for Natural Disasters"

This paper, which won the IFSA Best Student Paper award, presents a system for assessing the risk of natural disasters, which takes advantage of a fuzzy probability concept. This concept is also called possibility-probability distribution, because the uncertainty of the probability in this approach is expressed by a family of confidence intervals in form of a possibility distribution. This system was applied to assess the earthquake risk in Istanbul Metropolitan Area and the results were compared with an alternative approach. The achievement of this paper is due to two main characteristics. Firstly, it increases the reliability of the risk assessment by employing a more appropriate consideration of various uncertainties, and secondly, it implements this concept in a relatively uncomplicated methodology that makes it appropriate for use in practice.
Born in 1977, Mr. Iman Karimi is a graduate of Tehran University with a B.S. in Civil Engineering and a M.S. in Earthquake Engineering. He began his doctoral research on Earthquake risk analysis at the Department of Structural Statics and Dynamics, RWTH Aachen University, in 2003 and is currently completing his Ph.D. thesis, aiming to submit it by the end of year 2005.

Zengchang Qin, "Prediction Trees Using Linguistic Modeling"

This award winning paper proposed a new tree-structured prediction model based on fuzzy labels. By testing on some benchmark problems, it showed that this model has comparable accuracies and much better transparency than other state-of-art prediction algorithms. Some useful applications presented in this paper include sunspot prediction and flood forecasting.
Mr. Zengchang Qin obtained his B.Eng. degree in control engineering, Department of Automation, Heilongjiang University, Harbin, China, in 2001. In October of the same year, he started his M.Sc. study in Machine Learning and Data Mining in the Department of Computer Science, University of Bristol, UK. One year later, he won the ORS (Overseas Research Studentship) and Bristol University Scholarship for his PhD research under the supervision of Dr Jonathan Lawry from the A.I. Group of the Department of Engineering Mathematics at the same university. His current Ph.D. research is using a new fuzzy label framework proposed by Lawry for intelligent data mining system designs. His research interests include machine learning and data mining, soft computing, modeling with words, agent-based computational economics and evolutionary computing.

Xiaomeng Wang, "Mining Fuzzy Frequent Item Sets"

Due to various reasons, transaction data often lack information about some items. This leads to the problem that some potentially interesting frequent item sets cannot be discovered, since the number of supporting transactions found by classical mining methods, which rely on exact matching, may be smaller than the user-specified minimum. Aiming at this problem, we introduced the notion of "fuzzy frequent item sets" to model such potential frequent item sets that get lost due to missing information in the data. The term "fuzzy" means that an item set may not be exactly found in all supporting transactions, but only approximately. For mining fuzzy frequent item sets, we proposed approximate matching based on transaction editing, more precisely, on the virtual inserting of missing items into transactions. The recursive elimination algorithm presented here relies on a step by step elimination of items from the transaction database together with a recursive processing of transaction subsets. This algorithm is very simple, works without complicated data structures, and allows us to find fuzzy frequent item sets easily.
Ms. Xiaomeng Wang received her Bachelors degree in Electronic Engineering from Hefei University of Technology, China, and the degree of Diplom-Informatikerin (Masters degree in Computer Science) from Otto-von-Guericke University of Magdeburg, Germany. Currently she is pursuing a Ph.D. at the Institute of Knowledge Processing and Language Engineering, Otto-von-Guericke University of Magdeburg.

top

XIE Lijue, "The Core of Capacities on Lattices"

The Elsevier Young Researcher Award winning paper is about how to define the core of a given fuzzy measure (or capacity, game). We know that convex games have non empty cores and that the core is non empty if and only if the game is balanced. Recently, capacities have been defined in a general way on lattice structures such as l-ary capacity, multichoice game, bi-capacity, bipolar game, etc. This paper aims at defining the core for such capacities and to find their properties.
Ms. XIE Lijue graduated in Mathematics at Hubei University in 2001, completed 4 years' courses in 3 years. During 2002-2003, she continued her studies at Rouen University. She received her Masters degree in Mathematics, Informatics, Human Sciences at Panth?on-Sorbonne University (Paris 1) in 2004. Currently continuing her Ph.D. studies with Prof. M. Grabisch, she is interested in Fuzzy Measure, Capacity, Game Theory, Decision Theory, Social Choice etc.

top

ISL DTMI BUTE: The Cradle of Fuzzy Research of Hungary

The abbreviation in the title stands for Intelligent Systems Laboratory at the Department of Telecommunications and Media Informatics, Budapest University of Technology and Economics in Hungary. The roots of ISL go back to 1976 when Prof. L. T. Kóczy, the Head of the Lab, joined the Institute for Telecommunication Electronics and introduced fuzzy logic and intelligent systems research. In the beginning the activities focused mainly on theoretical subjects, especially algebraic structures for fuzzy connectives. However, possible applications in modeling, control, image recognition, and other areas were always kept in mind. Later the Lab's activities shifted towards more practical targets, when several international cooperation projects were firmly established and various national grants were obtained. Besides the permanent staff and graduate students, there are regularly a score of visiting researchers and short time visiting professors staying at the Lab. The main international contacts are with Prof. K. Hirota and others in Japan (under the framework of the Japanese - Hungarian Joint Research Laboratory), Prof. T. D. Gedeon and his group in Australia, Prof. A. E. Ruano and students in Portugal, and Prof. E. P. Klement and colleagues in Austria, with some further links being built to the Czech Republic and Norway. From the late 1980s Prof. K;oacute;czy's interests turned to engineering related problems, first the construction of extended fuzzy flip-flop circuits and then toward interpolation in fuzzy rule bases. The latter research lead to one of the major results of the group, KH-interpolation often referred in the literature under the name of the authors. In the 1990s a further step extended interpolation to hierarchical fuzzy systems and sub-models with different dimensionality. In the second half of the 1990s another main stream of research was started, the development of automatic identification methods for both simple ('flat') and hierarchical rule systems. In order to investigate alternative identification approaches, the fields of interest were extended to include clustering, evolutionary algorithms, and neural network based methods including the Levenberg-Marquardt and other competitive approaches. The most recent advancement, the bacterial memetic algorithm introduced at IFSA 2005 in Beijing, was obtained by combining a bacterial evolutionary algorithm and Levenberg-Marquardt techniques. In addition to these application related topics, the mathematical properties of fuzzy systems, such as approximation and complexity issues, have been extensively studied. Research areas of the Lab include singular value decomposition and tensor product based modeling and model reduction. Ongoing projects include an intelligent telecommunication supervision system, a fuzzy traffic control system, an agricultural field chemical concentration estimation model, an intelligent multimodal logistic hub, various linguistic information retrieval approaches, and the development of techniques to retrieve and classify Hungarian popular beliefs and superstitions.
Research in the Lab is extending the applicability of fuzzy systems and models to more and more complex practical problems and combining fuzzy logic with all aspects of soft computing and computational intelligence.

László T. Kóczy M. Sc. E. E. (1975); M. Phil. Control E. (1976); Ph.D. (1977), Technical University of Budapest, Doctor of the Hungarian Academy of Science, (1998, the highest earnable postdoctoral degree in Hungary) Visiting positions: Australian National University (Canberra); Murdoch University (Perth, Australia), University of New South Wales (Sydney, Australia); J. Kepler Universität Linz (Austria); University of Trento (Italy); Tokyo Institute of Technology (Yokohama, Japan; Chair Professor in 1993/94); Pohang Institute of Science and Technology (Korea). Summer University lecturing: Helsinki University of Economics, University of Helsinki (Finland), University of Minas Gerais (Belo Horizonte, Brazil), Dalian Maritime University (China). Research interests: Telecommunication systems, Intelligent models and systems, Very large and complex systems and networks. Professional Societies: International Fuzzy Systems Association: President 2001-2003; President Elect, 1999-2001; Vice President, 1995-99. IEEE: Senior Member, 1999-; NNS Regional Activities Chair 2001-2002; Chapters Chair 2002-.EURO WG on Fuzzy Sets (currently EUSFLAT): Founding Member, 1975-. Hungarian Fuzzy Society: Founding President 1990-1999, now Life Honorary President. He has published 326 papers including some textbooks.

top