INTERNATIONAL FEDERATION OF CLASSIFICATION SOCIETIES NEWSLETTER Number 22, December 2001 President: Jean-Paul Rasson Editor: Paul De Boeck ----------------------------------------------------------------- CONTENTS From the president IFCS-2002 Conference News from the IFCS News from the CSNA Conferences and meetings: CSNA Meeting 2001; SFC Meeting 2001; SFC Meeting 2002; GfKl Meeting 2002; GfKl AG-BT; Workshop Symbolic Data Analysis; Symposium on Clustering and Classification; Workshop on Developments and Challenges in Mixture Models ----------------------------------------------------------------- The International Federation of Classification Societies, founded in 1985, is composed of: Associaçáo Portuguesa de Classificaçào e Análise de Dados, British Classification Society, Central American and Carribean Society of Classification and Data Analysis, Classification Society of North America, Gesellschaft für Klassifikation, Irish Pattern Recognition and Classification Society, Japanese Classification Society, Korean Classification Society, Société Francophone de Classification, Società Italiana di Statistica, Vereniging voor Ordinatie en Classificatie and Section of Classification and Data Analysis of the Polish Statistical Society. The IFCS is a non-profit, non-political scientific organization, the aims of which are to further classification research. Among other activities, the IFCS organises a biennial conference, and supports the Journal of Classification. ----------------------------------------------------------------- FROM THE PRESIDENT Time is flying. My term of office as president of the IFCS will be over at the end of December and I realize that so many things are still to be done or even to be initiated. Nevertheless I think that the IFCS has developed a lot and extended its role as one of the major centers of scientific development and cooperation in the domain of classification and data analysis. Before all, this is the work of its worldwide membership in many different classification societies and groups. On the other hand, it would not have been possible without the activity of the members of the IFCS Executive Committee. So I would like to thank very warmly all the colleagues who have served in this Committee during these last two years. They have done an excellent job in a very constructive way and have resolved many practical and organisational problems. I also appreciated the fact that in my privileged position, I had the opportunity to contact so many members of the IFCS Member Societies and I was glad to do so and to find many friends and good ideas for the future. Also there was an unanimous feeling that the "visible part of the iceberg" of workload for IFCS, i.e., the IFCS conferences, have been great, scientifically attracting and splendidly organized. (Note that I do not judge my own role in the Namur conference, which was only one of these conferences, but I only cite was has been observed by others and what I observed in other conferences). At the moment these conferences render the IFCS a very important and well recognized scientific federation in the world. The IFCS and its conferences have evolved within time. A long time ago the rules for running the IFCS business and for organizing the conferences have been established in the form of by-law and guidelines. Also the Travel Award Program (TAP) and the corresponding TAP have been introduced. Given the development inside the IFCS, the growing size of IFCS Member Societies and IFCS conferences, and also the increasing number of upcoming tasks it seems to me that some of these guidelines have lost their importance or relevance whereas other rules prove to be unsufficient or incomplete in order to guarantee a smooth or clear functioning of business under the modified situation. So, at the end of my terms of office, I have the feeling that some of these rules should be updated in a professional way in order to meet the needs of a modern time and of a large and expanding Federation.For example, considering the number of Member Societies, I would like to suggest to adapt what is needed, for example, in the By-Laws. Or to modify the traditional organization of the Council meetings which take usually place before a Council dinner during the IFCS conference. With two Council assignments per Society, the additional members and the Excecutive Committee, this is now a meeting with, roughly, fourty people, trying to discuss IFCS matters and to take useful decisions for the Federation for the two years to come ­ all within two hours which is certainly not enough. I am sure that many colleagues share my opinion that we have to find a way to give back to the Council the important role it has in the Constitution and therefore provide better conditions of work than those provided by our tradition. Probably, there will be other rules or functions which have to be adapted in order to keep a high standard for our Federation. It is a challenge to consolidate its quality an to confront new horizons. Also there might be new work for the scientific activities in the IFCS ­ the two-years period between two conferences might be too long as to maintain contacts and foster cooperation in a sufficient way. So we should think about other ways of cooperation between IFCS Members Societies (which is already going on) and their individual members. I am sure that my successor in the office of IFCS President, Professor Carlo Lauro, will address some of these problems and I would like to congratulate him for having the opportunity to resolve them, together with the Council, in a splendid manner and thereby to reshape IFCS in order to cope with the future. I look forward to meeting many IFCS members in Krakow. Jean-Paul Rasson ----------------------------------------------------------------- IFCS-2002 CONFERENCE The forthcoming IFCS conference will take place in Cracow, Poland, on July 16-19, 2002. The meeting will be hosted by the Cracow University of Economics. As usually, the topics of the conference program belong to the area of classification, data analysis and related methods. Both theoretical and applied issues will be covered. The particular topics are as follows (but are not restricted to): General Topics: probabilistic methods, Bayesian data analysis, data mining, mixture models, internet surveys, graphs, validation, computational methods, combinatorial algorithms, information retrieval, neural networks Classification: decision trees, phylogenetic methods, pattern recognition, fuzzy clustering, hierarchical methods, discriminant analysis, optimisation in classification, constrained classification, nonhierarchical methods Data Analysis: textual data analysis, multidimensional scaling, robust data analysis, multivariate data analysis, time series analysis, regression trees, categorical data analysis, symbolic data analysis, correspondence analysis, principal component analysis, factor analysis, multiway data analysis, spatial data analysis Application of Classification or Data Analysis in: social sciences, behavioural sciences, finance, management, marketing, environmental sciences, ecology, biology, industry, medicine, genome analysis, archeology, image analysis, risk analysis, cognition, quality control. Researchers wishing to present a paper (as oral presentation or poster) should send an abstract (up to 200 words) together with a title, names and affiliation of authors plus keywords to the Chair of the Scientific Program Committee, Krzysztof Jajuga, before February 15, 2002. By default it is assumed that the author wishes to present the paper orally. However, if there are too many oral presentations some papers may be deferred to the poster session. In case the author prefers a poster presentation, he/she should specifically indicate this wish under the abstract. If there is more than one author, please indicate who will present the talk. Each person is permitted to present only one contributed paper. This does not prevent an invited session speaker from presenting the contribution nor does it limit the number of papers on which one may be listed as the co-author. The acceptance of the abstract will be notified before the end of March 2002. The conference language is English. The conference fee is 250 EURO. The fee for students is 150 EURO if a copy of a student registration card is provided. A reduction of 50 EURO applies if payments are received before March 1, 2002. Useful addresses: Local Organizing Committee IFCS-2002, Andrzej Sokolowski, Cracow University of Economics, ul. Rakowicka 27, 31-510 Kraków, Poland; phone:+48 12 2935003, fax: +48 12 2935004, e-mail: ifcs2002.ae.krakow.pl Scientific Program Committee IFCS-2002, Krzysztof Jajuga, Wroclaw University of Economics, ul. Komandorska 118/120, 53-345 Wroclaw, Poland; phone: +48 71 3680340, +48 71 3680338, fax: +48 71 3680322, e-mail: jajuga@manager.ae.wroc.pl World Wide Web: http://ifcs2002.ae.krakow.pl Hotel and social events reservation: First Class Travel Office, Zwierzyniecka 29, 31-105 Krakow, Poland, phone/fax: +48 12 4311385 Note that a workshop on symbolic data analysis will be organized the day before the conference, see p. 7. ----------------------------------------------------------------- NEWS FROM THE IFCS At the end of this year, IFCS President Professor Jean-Paul Rasson will step down, to be replaced by Professor Carlo Lauro. Jean-Paul will continue to serve as Vice-President until the end of 2002. Under Jean-Paul's leadership, the IFCS held its millenium conference in Namur, established a standing program to award travel funding to young researchers in the classification sciences, and was an invited participant in the Seoul meeting of the International Statistical Institute. Jean-Paul has also played a major role in preparing and organizing the 2002 IFCS meeting, which will be held in Krakow, Poland. The entire IFCS owes Jean-Paul a debt of gratititude for his statesmanship, his dedication, and his vision. I hope that everyone is planning to attend the 2002 meeting of the IFCS. The local organizers and the program committee are working hard to make it a great success. Please bear in mind that the deadline for contributing an oral paper is February 15, 2002 (the deadline for submitting a paper for the conference proceedings has already passed). Additional details can be found by contacting the conference website at: http://ifcs2002.ae.krakow.pl The IFCS Travel Awards Program, under the chairmanship of Chikio Hayashi, is planning to provide partial funding to selected young researchers to support attendence at the Krakow meeting. Each member society is invited to nominate a few good people under the age of 35 (some flexibility on the age limit may be obtained in exceptional circumstances). Details on the nomination procedure are provided on the IFCS website at: http://www.classification-society.org and the Awards Committee hopes to receive all nominations by late January. David Banks IFCS Secretary ----------------------------------------------------------------- NEWS FROM THE CSNA Meeting 2001 CSNA's Annual Meeting for the year 2001 took place from June 14 to 17 at the Washington University School of Medicine, St. Louis, Missouri, hosted by Professor William D. Shannon. Shannon is a Biostatistician and Director of the Department of Medicine Biostatistics Consulting Center at the School of Medicine. The medical campus is located at St. Louis' Central West End Metrolink and is directly across from beautiful Forest Park. Two short courses were held on June 14. These were "Using Latent Class Models for Clustering and Classification, " by Jay Magidson of Statistical Innovations, and "Axiomatizations of Consensus Problems in Classification and Group Choice," by William H.E. Day. The regular session began on the morning of Friday, June 15th. The plenary address was given by Prof. Dr. Willem Heiser, University of Leiden, The Netherlands, on "Early Statistical Modeling of Latent Quantities: The History of Distance Measurement by Triangulation." The President's Invited Session, organized by CSNA President David Banks, featured Karl W. Broman of Johns Hopkins University speaking on "Inference of Clusters of Relatedd Individuals," Tae Rim Lee, University of Wisconsin - Madison, "Tree Structured Prognostic Model for Hepatocellular Carcinoma," and David Hunter, Pennsylvania State University, on "MM Algorithms and Bradley-Terry Models." There was an Invited Address by Giovanni Parmigiani of Johns Hopkins on "Screening Genes for Expression-based Molecular Classification."A listing of all titles and authors will appear in the Journal of Classification. There was an excellent mix of theoretical and applied work on classification in different fields. This year's meeting was no exception to the fact that CSNA meetings are traditionally informal and very interdisciplinary. Abstracts of papers are distributed, but no formal proceedings are produced. Speakers are encouraged to discuss work in progress, of either applied or methodological nature. There was a reception on Thursday evening. The conference banquet on Friday evening was held at the International Bowling Museum and Cardinal Baseball Hall of Fame. The general business meeting took place on Saturday afternoon. Meeting 2002 CSNA's annual meeting for 2002 will take place at the University of Wisconsin, Madison, from June 13 to 16, 2002, hosted by Professor Bernie Harris of the Department of Statistics at Madison. There will be a number of excellent invited talks. Persons wishing to contribute a paper should contact Professor Harris via harris@stat.wisc.edu before 15-March-2002. Elections Within a few weeks, eligible voters will receive ballots for the annual election. This year, there is an election for President-Elect, the candidates being Phipps Arabie and Bill Shannon. Every year, the terms of two Board members end, and two new members are elected. The candidates this year are David Dubin, Herbie Lee, Mark Rorvig, and Kert Viele. Journal of Classification On the most recent (2000) ranking by ISI, the Institute for Scientific Information, JoC is 18th of 156 math journals. Its Impact Factor is .846 and the half-life of an article 8.4 years. Phipps Arabie is stepping down after eighteen and a half years of excellence as Founding Editor. Willem Heiser will assume the Editorship in March, 2002. As many IFCS members know, Willem will bring to the job a wealth of statistical, scientific and editorial experience, including a term as editor of Psychometrika. Arabie will continue as Acting Editor for several months on manuscripts submitted prior to March 1, 2002. Beginning on that date, all submissions should go to Prof. Willem Heiser, Leiden Univ., Faculty of Social & Behavioral Sciences, PO Box 9555, 2300 RB Leiden, The Netherlands, heiser@rulfsw.fsw.leidenuniv.nl Dr. Jennifer Pittman of Duke University's Institute for Statistics and Decision Sciences (ISDS) is the new Editor of the CSNA Newsletter. Dr. Pittman was formerly at the National Institute of Statistical Sciences (NISS). Editions of the CSNA Newsletter are available at: http://www.niss.org/pittman/csna/csnanews.html ----------------------------------------------------------------- CONFERENCES AND MEETINGS SFC MEETING 2001 SFC’2001 (the 8th meeting of the Société Francophone de Classification) will be held on the campus of the university of French West Indies at Pointe-à-Pitre, Guadeloupe (F.W.I), 16-21 December 2001. The conference will convene on Monday, 17 December. The program will include 25 sessions and 5 invited sessions. The program will also include tutorial sessions or short courses. Below is the temporary scientific program of SFC’2001. Sunday, 16 December 2001 16h00 - 18h00 Welcoming - Registration Monday, 17 December 2001 08h00 - 08h30 Welcoming - Registration 08h30 - 09h15 Inaugural session 09h15 - 10h15 Inaugural talk: M. COTRELL (Invited talk) 10h30 - 12h35 Session 1: Learning - Classification and neuronal methods 14h00 - 15h40 Session 2: Combinatory methods, graphs, optimisation 16h10 - 17h25 Session 3: SAS session 18h00 - 19h00 Welcome Cocktail Tuesday,18 December 2001 08h30 - 09h30 O. GASCUEL (Invited talk) 09h35 - 10h25 Session 4: Shape recognition 10h50 - 12h30 Session 5: Combinatory methods, graphs, optimisation 14h00 - 15h40 Session 6: Mathematical and statistical methods 16h10 - 17h10 Session 7: Round table on Classification Wednesday,19 December 2001 08h30 - 09h30 P. LEGENDRE (Invited talk) 09h35 - 10h25 Session 8: SODAS session 1 Session 9: Poster session 10h50 - 12h30 Session 10: SODAS session 2 Session 11: Mathematical and statistical methods 14h00 - 15h40 Session 12: SODAS session 3 Session 13: Combinatory methods, graphs, optimisation 16h10 - 17h25 Session 14: SODAS session 4 Session 15: Image analysis and signal analysis 17h30 - 18h30 Annual general meeting of the SFC Thursday,20 December 2001 08h30 - 09h30 F. NICOLAU (Invited talk) 09h35 - 10h25 Session 16: Mathematical and statistical methods Session 17: Free session on Non symmetrical data analysis 10h50 - 12h30 Session 18: Free session on Non symmetrical data analysis Session 19: Symbolic/Numerical approach 14h00 - 18h00 Social program Friday,21 December 2001 08h30 - 09h30 Simon Régnier Prize winner talk, prize giving 09h35 - 10h25 Session 20: Combinatory methods, graphs, optimisation Session 21: Applications 10h50 - 12h30 Session 22: Poster session Session 23: Algorithms and Classification 14h00 - 15h40 Session 24: Free session on symbolic/ numerical approach Session 25: Applications 16h10 - 17h10 Closing talk: H. BOCK (Invited talk) 17h10 - 17h30 Closing session 19h00 - 23h00 Closing party For more informations: http://sfc2001.univ-ag.fr ----------------------------------------------------------------- SFC MEETING 2002 9th annual Meeting of the SFC Toulouse, September 16-18 2002 After the 8th Conference in Guadeloupe in December 2001, this meeting will be jointly organized in Toulouse by 2 universities and 3 laboratories: The team GRIMM (statistics and graph theory) of the Mirail University of Social and Human Sciences, and the TCI (images processing) and LSP (statistics and probability) laboratories of the Sabatier University of Sciences. Neural Networks, Graphs, Image Processings, Multivariate Modelling, Pattern Recognition and Data Mining will be some important themes of the Meeting, with other various and usual thematic sessions of Classification. The deadline for submitting the short papers is March 1. More information on the internet: http://www.irit.fr/SFC2002/ Contact: sfc@cict.fr ----------------------------------------------------------------- GfKl MEETING 2002 The German Classication Society GfKl (Gesellschaft für Klassikation), a member of IFCS (International Federation of Classication Societies), will hold its 26th Annual Conference under the title: "Between Data Science and Everyday Web Practice" The scientic program will include plenary and semi-plenary lectures as well as contributed paper sessions on special topics. It is, this year, organized with strong support of members from CLADAG, the Classication and Data Analysis Group of the Società Italiana di Statistica. The conference will emphasize interdisciplinary research and the interaction between theory and practice. The broad range of relevant topics from classication, data analysis, and innovative applications is illustrated by the following list of sessions (organizers in brackets): - Classication: discrimination and clustering (Bock, Ritter) - Analysis of mixtures (Ambroise, Govaert) - Probability models in cluster analysis (Godehardt, Hennig) - Computational learning theory for classication (Obermayer) - Clustering of complex data (Vichi) - Dissimilarities and clustering structures (Rasson) - Non-parametric and non-linear regression and classication (Monari) - Exploratory data analysis (Critchley) - Analysis of symbolic data (Lauro) - Multiway data analysis (Rocci) - Computational advances in data analysis (Weihs) - Visualization and scaling methods (Blasius) - Data mining (Coppi) - Data analysis for structuring web search strategies (Säuberlich) - Web content and usage mining (Schmidt-Thieme, Spiliopoulou) - Virtual learning and new media (Schader) - Recommender systems and eCRM (Gaul, Geyer-Schulz) - Electronic commerce and administration (Oechsler) - Market research, marketing, and retailing (Baier, Decker) - Finance, capital markets, and risk management (Bühler) - Production, logistics, and controlling (Mißler-Behr, Opitz) - Linguistics and information retrieval (Goebl, Krause) - Subject indexing and library science (Hermes) - Data analysis in archeology (Ihm) - Bioinformatics (Vingron) - Genome and DNA analysis (von Haeseler) - Medicine and health sciences (Wernecke) - Analysis and classication of sensorial data (D'Ambra) - Neural data analysis (Oja) - Neural networks and genetic algorithms (Battaglia) - Analysis of spatial/temporal data (Fahrmeir, KnorrHeld) The conference will take place in the lecture halls of the University of Mannheim. The University is located in the Mannheimer Schloß, within five minutes walking distance from the Hauptbahnhof (Central Railway Station) in Mannheim. The easiest way to reach Mannheim from international destinations is to fly to Frankfurt/Main Airport which has a direct link by German Railway to Mannheim. Trip length is approx. 30 minutes. A Proceedings volume will be published in English by Springer-Verlag. Interested persons can obtain further information at: http://www.gfkl.de/gfkl2002/ and from the organizers. Online registration is possible. Participants intending to present a talk are requested to submit an abstract by March 15, 2002 to the local organizer, Martin Schader (Mannheim). Accepted abstracts will be included in a summary volume that will be distributed to conference participants. Speakers will be notied of acceptance by May 15th, 2002. Upon acceptance, a complete manuscript for the Proceedings can be submitted. It will have to pass a review process. Formatting instructions for abstracts and manuscripts can be obtained from: http://www.gfkl.de/gfkl2002/ ----------------------------------------------------------------- GfKl AG-BT ´´Classification and data analysis in biology and medicine´´ Workshop Proteomics, Bioinformatics and Genomics (5.-6. October 2001, Conference centre DELOS, Bernried at the ´Starnberger See´, Germany) The special interest group "Classification and data analysis in biology and medicine" (AG-BT) of the GfKl organized a joint workshop on "Proteomics, Genomics and Bioinformatics" together with the group "Functional genomics" of the Vereinigung für allgemeine und angewandte Mikrobiologie (VAAM) (Society of general and applied microbiology). About 50 scientists participated in the successful workshop organised by H.-P. Klenk (VAAM/GfKl) and B. Lausen (GfKl). Invited talks were on: ´´Protein interaction in the model organism yeast´´ (P. Uetz, Univ. Karlsruhe), ´´A primer in proteomics´´ (H. Platsch, BASF Mannheim) and ´´The HUSAR bioinformatics service´´ (M. von der Linden, DKFZ, Heidelberg). Other papers were on: ´´On the transcriptom of Corynebacterium glutamicum´´ (O. Sorgenfrei, BASF, Mannheim), ´´The upcoming genome project of Solrangium cellulosum Soce56´´ (S. Beyer, GBF, Braunschweig), ´´C. glutamicum Proteomics´´ (J. Kalinowski, Univ. Bielefeld), ´´Genome-wide operon mapping´´ (U. Zuber, Univ. Greifswald), ´´Analysis of transcriptional regulation with DNA-microarrays for two haloarchaeal species´´ (J. Soppa, Univ. Frankfurt), ´´GenDB - eine open source Genomannotations-Software´´ (V. Meyer, Univ. Bielefeld), ´´A bioinformatic toolbox for enhanced proteomic data analysis´´ (C. Garcia-Rizo, MPI Biochemie, Martinsried), ´´QUARCS - Quality recording and confidence evaluation of sequencing data´´ (A. Zibat, Epidauros AG Bernried), ´´Analyzing microarray gene expression data with biostatistical methods´´ (G. Welzl, LMU München), ´´Data driven modelling of the dynamics of a cellular signalling pathway´´ (J. Timmer, Univ. Freiburg), ´´Prognostic modelling with gene expression data´´ (B. Lausen, Univ. Erlangen), ´´Text mining of biological texts´´ (C. Gieger, LION Biosciences, Heidelberg), ´´Tumor classification by gene expression data´´ (M. Beibel, Univ. Freiburg), ´´Estimating Parameters in Dynamical Systems´´ (W. Horbelt, Univ. Freiburg), ´´Fast phylogeny reconstruction and application in biomedicine´´ (G. Füllen, Univ. Münster). ----------------------------------------------------------------- STATISTICAL COMPUTING 2002 (23.-26.6.2002, Schloß Reisensburg) In June 2002 the AG-BT of the GfKl, the group ''Statistische Auswertungssysteme'' der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e.V. (GMDS) and the group ''Computational Statistics'' of the German region of the International Biometric Society organise the meeting ''Statistical Computing'' on methods and tools of computer science for statistics at Schloß Reisensburg, Günzburg, Germany (Local organiser: Axel Benner). Bioinformatics is the special topic. Other Topics are: Web-based ''Statistical Computing'', Markov Chain Monte Carlo methods, statistical graphics, pattern recognition, Bayesian-methods and data mining. Further Information: http://www.dkfz-heidelberg.de/biostatistics/ Reisensburg2002/ ----------------------------------------------------------------- Workshop Symbolic Data Analysis Cracow (Poland) - Monday, 15th July 2002 SCIENTIFIC PROGRAM COMMITTEE - International Members H. Bacelar-Nicolau (Portugal), V. Batagelj (Slovenia), L. Billard (USA), H-H. Bock (Germany), C. Bravo (Spain), P. Brito (Portugal), F.A.T. De Carvalho (Brazil), E. Diday (France), M. Chavent (France), F. Esposito (Italy), M. Gettler-Summa (France), K.S. Gowda (India), A.Hardy (Belgium), M. Ichino (Japan), C. Lauro (Italy), Y. Lechevallier (France), D. Malerba (Italy), C. Marcelo (Portugal), M.Noirhomme (Belgium), H. Papageorgiou (Greece), J.P. Rasson (Belgium), O. Rodriguez (Costa-Rica), R. Verde (Italy), S. Winsberg (USA). ORGANIZING COMMITTEE Rosanna Verde, Second University of Naples, Italy (chair) Paula Brito, University of Porto, Portugal Francisco de A.T. de Carvalho, Federal University of Pernambuco, Brazil Andrzej Sokolowski, University of Cracow, Poland (local organizer) TOPIC Symbolic Data Analysis (SDA) is a new domain in the knowledge discovery and management area, related to multivariate analysis, pattern recognition, data bases, artificial intelligence. As input the aim of SDA is to allow a more realistic description of the units by taking care of their internal variation and their complex structure. As output the aim of SDA is to provide a better explanation of the results by an automatic interpretation closer to the user natural language. SDA provides, suitable tools to work with complex, aggregated, relational and higher-level data described by multi-valued variables where the entries of a data table are sets of categories or of numbers, intervals or probability distributions, related by rules and taxonomies. The methodological issues under development generalise the classical exploratory data analysis techniques, like: visualisation, factorial techniques, decision tree, discrimination, regression, multidimensional scaling, classification and clustering methods, conceptual lattices. The aim of this workshop is to present new theoretical results and applications in this framework. Researchers interested in SDA are invited to submit an abstract by e-mail to: rosanna.verde@unina2.it in PDF format (please send it as .ZIP file). (max 2 pages - following the guidelines that can be found at the IFCS2002 web site: http://ifcs2002.ae.krakow.pl) Deadlines and Time schedule: - March 1st, 2002 - Deadline submission abstract - March 31th, 2002 - Notification of the acceptance of the abstract ----------------------------------------------------------------- SYMPOSIUM ON CLUSTERING AND CLASSIFICATION January 11-12, 2002 The Department of Statistics of the University of Florida will host its Fourth Annual Winter Workshop on January 11-12, 2002 in Gainesville, Florida. The topic of the workshop is classification and clustering. The symposium is organized by George Casella, Sam Wu, Jim Booth, Dave Wilson, Rongling Wu, Brett Presnell, Alan Hutson and Jim Kepner. Invited speakers will include Christopher Small (U. of Waterloo), Edward George (U. of Pennsylvania), Peter Hall (Australia), Nick Lange (Harvard U.), Robert Paige (Texas Tech U.), Dennis Pearl (Ohio State U.) Anuj Srivastiva (Florida State U.), Keith Worsley (McGill U.) Mark Yang (U. of Florida), Hongyu Zhao (Yale U.), Christopher Genovese (Carnegie Mellon U.), Art Owen (Stanford U.), and Terrence Fine (Cornell U.). In addition to invited papers, the symposium will include a contributed poster session. The graduate school at the University of Florida has contributed funds for the purpose of supporting a small number of graduate students from peer institutions to attend the symposium. Each student will receive free lodging at the Reitz Union Hotel for up to three nights, in addition to a registration waiver which includes two breakfast, one lunch and a BBQ. Students are expected to share rooms, two to a room. Interested students should contact: icc@stat.ufl.edu For more information: Carl Rozear, Dept. of Statistics, University of Florida, P.O. Box 118545, Gainesville, Florida 32611-8545 USA, phone: (352)-392-1941 ext. 207, fax (352) 392-5175. ----------------------------------------------------------------- WORKSHOP ON DEVELOPMENTS AND CHALLENGES IN MIUXTURE MODELS, BUMP HUNTING AND MEASUREMENT ERROR MODELS June 2 to 4, 2002, Cleveland, Ohio PRELIMINARY ANNOUNCEMENT This special international research workshop brings together researchers from different fields such as astronomy, biology, bioinformatics, data mining and statistics, but who are all interested in the scientific challenges to their fields arising from the statistical problems associated with mixture models, bump hunting, measurement error models and other related topics. The workshop is designed to provide a unique opportunity for young researchers and experts alike to closely interact on these topics. Contact workshop organizers for possible travel support to junior researchers and under-represented groups. Invited Sessions: - Astronomy - Bayesian computational methods - Bioinformatics - Bump hunting - Clustering - Genetics - Image analysis - Measurement errors - Mixture models - Neyman lectures For more details, see web site: http://sun.cwru.edu/mix ----------------------------------------------------------------- IFCS NEWSLETTERS ON INTERNET: The newsletters are available as PDF files on the IFCS website (see below). The files can be read with AcrobatReader. Click on Acrobat Reader to download. IFCS HOMEPAGE: www.classification-society.org The website is designed and maintained by David Dubin. The website contains among others: the IFCS Constitution and By-Laws, the IFCS newsletters, and pointers to the websites of the member societies. The URL is new. The old URL will continue to redirect browsers to the new site. ----------------------------------------------------------------- NEWS FOR THE COMING ISSUE OF THE IFCS CAN BE SENT TO: Paul.Deboeck@psy.kuleuven.ac.be Text files are by preference in ascii, wordperfect, or word, with pc format (mac is difficult, although not impossible). For graphical materials, by preference BMP, GIF, or PCX are used. Editor: Paul De Boeck, Department of Psychology, University of Leuven, Tiensestraat 102, B-3000 Leuven, Belgium. tel: +32 16 325980 fax: +32 16 215916