KDD Nuggets 95:28, e-mailed 95-11-03 Contents: * M. Holsheimer, Data Distilleries Company, http://www.ddi.nl * GPS, Intelligent Agents in Netscape * A. Tuzhilin, CFP: Special Issue of DSS on Knowledge Discovery Siftware: * J. Betak, DOS Port of AutoClass C Bayesian Classifier version 2.7 * Zighed, Machine learning program for PC/Windows Meetings: * J. Talavera, IPMU'96: Information Processing and Management of Uncertainty, Granada, Spain, July 1-5, 1996, http://pirata.ugr.es/ipmu96.html -- The KDD Nuggets is a moderated mailing list on Data Mining and Knowledge Discovery in Databases (KDD). Please include a DESCRIPTIVE subject line and a URL, when available, in your submission. Nuggets frequency is approximately weekly. Back issues of Nuggets, a catalog of S*i*ftware (data mining tools), references, and other related information is available at Knowledge Discovery Mine, URL http://info.gte.com/~kdd E-mail add/delete requests to kdd-request@gte.com E-mail contributions with a DESCRIPTIVE subject line to kdd@gte.com. -- Gregory Piatetsky-Shapiro (moderator) ********************* Official disclaimer *********************************** * All opinions expressed herein are those of the writers (or the moderator) * * and not necessarily of their respective employers (or GTE Laboratories) * ***************************************************************************** ~~~~~~~~~~~~ Quotable Quote ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Where is the wisdom we have lost in knowledge? Where is the knowledge we have lost in information? -from "The Rock" by T.S. Eliot (thanks to Haym Hirsh) >~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ To: kdd@gte.com Subject: Data Distilleries founded http://www.ddi.nl Date: Wed, 01 Nov 1995 22:46:50 +0100 From: Marcel Holsheimer Data Distilleries: putting data mining technology to practice Data Distilleries is the first company whose core business is the development of data mining software. Data mining is an analysis technique, used to discover hidden strategic information in very large databases. This is information that cannot easily be revealed using conventional analysis techniques, such as statistical tools. Data Distilleries (DD) offers advanced data mining tools, but also consultancy and training to enable users to get the most out of these tools. One of the products of DD is `Data Surveyor'. This is an advanced data mining tool that enables the users to search their own databases, and present the discovered information in an easy to understand, graphical format. Data Surveyor can be coupled with the user's databases. Dutch banks and insurance companies use DD's expertise and techniques to discover risk profiles in their customer databases. Or they use data mining techniques to select the most interesting customers for direct marketing purposes. DD uses these techniques for many other purposes as well, such as determining which combinations of machine and environmental factors cause faulty products in a mass production process. DD was founded in September as a spin off of CWI, the Dutch research centre for mathematics and computer science. CWI is one of the leading European institutes in the area of data mining and has extensive experience in the development of data mining tools and their application to real-world problems. Being the project leader of the EC funded KESO project, DD has intimate relations with other top European institutes, such as the University of Helsinki and GMD in Germany. More information on DD can be found on the internet: http://www.ddi.nl Or by contacting us at: Data Distilleries Kruislaan 419 1098 VA Amsterdam The Netherlands Tel +31 (0)20-560 8433 fax +31 (0)20-668 5486 Email: marcel@ddi.nl >~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Date: Fri, 27 Oct 1995 14:19:45 -0400 From: gps@gte.com (Gregory Piatetsky-Shapiro) Subject: Intelligent agents in Netscape David Blanchard reports in Applied AI News page in AI Magazine, Fall 1995, issue, that Verity (Mountain View, CA) company will provide their "Topic" intelligent agent technology to Netscape. Netscape will embed Verity's Topic Search Engine in its servers, and allow users to filter incoming information against user profiles and send automatic alerts via personal HTML pages, e-mail, or fax. >~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Date: Mon, 30 Oct 95 15:35:07 EST From: atuzhili@square1.stern.nyu.edu (Alex Tuzhilin) Subject: CFP: Special Issue of DSS Journal on Knowledge Discovery Call For Papers DECISION SUPPORT SYSTEMS JOURNAL Special Issue on Knowledge Discovery and Its Applications to Business Decision Making Over the past two decades, corporations have collected massive volumes of data about their day-to-day operations, their customers, competitors, employees, and other useful data. In addition to the explosive increase in volume, the structure and complexity of the data in many applications, such as market research, financial, and telecommunication applications, has grown at an incredible rate as well. Instead of working with one or two data sets, these new applications require dozens of different data sets, some of them having over a hundred variables with complex relationships among them. This data, if managed and ``harvested'' properly, can be an extremely valuable asset to organizations. Unfortunately, the explosive growth of data has often resulted in frustrating situations when organizations were unable to understand well what information the data contained and thus could not leverage the data. To address this problem, the field of Knowledge Discovery in Databases (also referred to as Data Mining) was formed and attracted considerable attention of statisticians, machine learning and database researchers, and practitioners in industry. SCOPE To address the growing interest in the knowledge discovery field, the DSS Journal is planning a special issue on Knowledge Discovery in Databases, scheduled to appear in June 1997 with the focus on business applications and on the relationship between knowledge discovery and business decision making. High quality original research papers are solicited on all aspects of Knowledge Discovery in Databases, as long as they can demonstrate applicability of their results to solving practical business problems. Topics of interest include, but are not limited to * Business Applications of Knowledge Discovery -- Successes and Failures; Lessons Learned from These Applications * Relationship Between Knowledge Discovery and Decision Support Systems * Methodologies and Frameworks of Knowledge Discovery * Knowledge Discovery Tools and Techniques and Their Applications in Business Of particular interest are the papers that will not only contribute to the academic arena, but will also benefit the business community in the foreseeable future and that would encourage a dialog between researchers and practitioners. Prospective authors are encouraged to submit both practical and theoretical papers. However, the authors of practical papers are encouraged to describe the lessons learned from the applications they developed and indicate how their methods can be extended to other types of applications. Similarly, the authors of theoretical papers are encouraged to demonstrate how their theoretical developments can be applied in business to solve practical problems. THE JOURNAL Decision Support Systems (DSS) is a journal published by Elsevier Science Publishers (North-Holland) since 1985. It publishes articles covering various technical aspects of DSS, including concepts and operational basis for DSSs, techniques for implementing and evaluating DSSs, DSS experiences, and related studies. Articles published in the journal delve into, draw-on, or expand such diverse areas as artificial intelligence, cognitive science, computer supported cooperative work, database management, decision theory, economics, linguistics, management science, psychology, user interface management systems, and others. The common thread of articles published in the journal is their relevance to technical issues of DSS. SUBMISSION AND REVIEW CRITERIA Four (4) copies of a manuscript and a single title page should be submitted by MAY 31, 1996 to the Guest Editor Alex Tuzhilin Information Systems Department Stern School of Business New York University 44 West 4th Street, Rm. 9-78 New York, N.Y. 10012 E-mail: atuzhili@stern.nyu.edu Tel: 212-998-0832 FAX: 212-995-4228 The length of the paper should not exceed 25 double-spaced, single-sided, 12-point-type pages (including figures and references). The title page should contain the title of the paper, the name(s) and institutional affiliation(s) of the author(s), an abstract of no more than 200 words, a list of keywords, and the name, address, telephone number, and e-mail address of the contact author. The papers will be refereed in accordance with the usual procedure of the DSS Journal (see ``Instructions to Authors'' at the end of any issue of the journal for further details about the submission format) and will be judged based on their relevance to the KDD field, originality, information content, validity, and readability. IMPORTANT DATES: Submission deadline: Friday, May 31, 1996 Acceptance notification: Friday, September 13, 1996 (!) Final manuscript due: Friday, November 15, 1996 Publication date of issue: June 1997 >~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ From: jbetak@rz.fh-augsburg.de (Betak Juraj) Subject: DOS Port of AutoClass C Bayesian Classifier version 2.7 To: kdd@gte.com Date: Sat, 28 Oct 1995 14:38:30 +0100 (MET) Dear Sir, AutoClass C now also runs on a DOS platform. Here I am forwarding a message from Will Taylor, with whom I worked on the port. Maybe it would be worth the effort to update the S*i*ftware WWW page :-) Best regards > Date: Tue, 24 Oct 95 09:27:07 PDT > From: taylor@ptolemy.arc.nasa.gov (Will Taylor) > Message-Id: <9510241627.AA19218@muir.arc.nasa.gov> > To: jbetak@rz.fh-augsburg.de > Cc: martin@informatik.fh-augsburg.de > Subject: DOS Port of AutoClass C Bayesian Classifier version 2.7 > Reply-To: taylor@ptolemy.arc.nasa.gov > Content-Length: 4650 > X-Lines: 115 > > Juraj - > This message was posted to the sci.math.stat and comp.ai usenet > newgroups and to comp.infosystems.www.announce. > > Thanks for all your diligent work on this port. > > ==> Will > > ------------------------------------------------------------------ > Announcing the DOS release of the port of version 2.7 of AutoClass C, > the Bayesian classifier which seeks a maximum posterior probability > classification. > > Key features: > - determines the number of classes automatically; > - can use mixed discrete and real valued data; > - can handle missing values; > - processing time is roughly linear in the amount of the data; > - cases have probabilistic class membership; > - allows correlation between attributes within a class; > - generates reports describing the classes found; and > - predicts "test" case class memberships from a "training" > classification. > > Inputs consist of a database of attribute vectors (cases), either real > or discrete valued, and a class model. Default class models are provided. > AutoClass finds the set of classes that is maximally probable with > respect to the data and model. The output is a set of class descriptions, > and partial membership of the cases in the classes. > > Summary of updates: > > Version: 1.0 = 14 Oct 95 Port of AutoClass C to > corresponds to AutoClass C UNIX/LINUX/IRIX > version 2.7. Note that it is in a separate distribution file: > AUTOCLAS.ZIP. > > ----------------------------------------------------------------------------- > The DOS port was done by Juraj Betak (jbetak@rz.fh-augsburg.de). > ----------------------------------------------------------------------------- > > This DOS port of the UNIX version of AutoClass has been done carefully > and to the best of my abilities. There is no warranty though that it > will perform as well as the original UNIX version. > > AUTOCLAS.ZIP was created by the pkzip utility, version 1.1 (03/1990), > and should be upward compatible to any more recent pkunzip version. > The executable form of the program (PKUNZIP.EXE) is obtainable from the > FTP site. To unpack all the files along with the directory hierarchy, > type: > 'PKUNZIP -D AUTOCLAS.ZIP' > > For information or help concerning the AutoClass program, see the contact > below. AutoClass is *not* copyrighted software. > > Please check csr.uta.edu:/pub/Readme, whether new version has been released > in the meantime. There might not exist a complete DOS port of the newest > version at this time, again please check the WWW site below or the FTP site. > > FTP site: csr.uta.edu:/pub/AUTOCLAS.ZIP (transfer in binary mode) > csr.uta.edu:/pub/PKUNZIP.EXE (transfer in binary mode) > > Alternate FTP site: > ftp.gmd.de:/incoming/AUTOCLAS.ZIP (transfer in binary mode) > ftp.gmd.de:/incoming/PKUNZIP.EXE (transfer in binary mode) > > -------------------------------------------------------------------------------- > > When you have trouble in obtaining GNU C via ftp, just turn to the SimTel > site at: > > ftp.coast.net:/SimTel/vendors/djgpp > or http://www.coast.net/SimTel/ (look for vendors collection) > > please, look for the basic files needed to run GNU C: > > gcc260dc.zip (Standard docs) > bnu24bn.zip > djdev112.zip > gcc260bn.zip (compiler binaries) > djeoe112.zip > gas23bn.zip (assembler) > > The numbers might differ, when a newer version has been released in the > meantime. > > GNU C is distributed under the GNU Software licence and is more or less free, > unless you intend to use it in commercial applications. > > The SimTel collection is one of the largest PC Software sources and it is > also very well maintained. SimTel has lots of mirrors around the worlid, > there is a good chance that you have one just "next doors" and don't have > to access a foreign site (in case your ftp access is limited). > > -------------------------------------------------------------------------------- > > The latest UNIX/IRIX/Linux version of AutoClass is available free of charge, > and may be obtained via anonymous FTP from the same site. > > FTP site: csr.uta.edu:/pub/autoclass-c.tar.Z (transfer in binary mode) > > Contact: Will Taylor > NASA Ames Research Center > MS 269-2, Moffett Field > CA 94035-1000 > U.S.A. > > taylor@ptolemy.arc.nasa.gov > +1 415 604-3364(voice) or 604-3594(fax) > > WWW page: http://ic-www.arc.nasa.gov/ic/projects/bayes-group/group/html/ > autoclass-c-program.html > > <<<------------------------------------------------------------------->>> > > >~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Date: Tue, 24 Oct 1995 20:26:26 +0100 X-Sender: zighed@diogene.univ-lyon2.fr To: ml@ics.uci.edu, kdd@gte.com From: zighed@univ-lyon2.fr (Zighed) Subject: Machine learning program for PC/Windows Dear users, We have the pleasure to offer you the first platform of "knowledge engineering" (SIPINA-W) ). YOU will thus be able to : 1- generate knowledge in the form of production rules and this, by learning from a basis of examples characterized by numeric and/or symbolic data. 2-test on non-learned examples knowledge produced before. 3-add knowledge generated automatically by SIPINA-W) to other knowledge which can be given by a human expert. 4-merge and optimize bases of knowledge produced artificially or by a human expert... SIPINA-W) offers you numerous options which allow you to use methods such as C4.5. You will find many applications made on known examples in medical, industrial or financial fields... You can have access to SIPINA-W) by: ftp.univ-lyon2.fr /pub/pc/Eric/SIPINA The version we present here is a beta test version. You can duplicate and diffuse this software freely. We only ask you to be so kind as to send us an e-mail to give us your address. We will thus be able to send you the new updatings. You can also indicate us mistakes we could have missed. Thank you and see you soon, Prof. D.A. Zighed Equipe de Recherche en Inginierie des Connaissances Universiti Lumihre Lyon 2 5 av. Pierre Mendhs-France 69676 Bron France tel. (33) 78 77 23 76 Fax. (33) 78 77 23 75 e-mail : zighed@diogene.univ-lyon2.fr >~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Date: Fri, 27 Oct 1995 09:35:39 GMT From: "J. Carlos Cubero Talavera" To: kdd@gte.com Subject: IPMU'96: Information Processing and Management of Uncertainty in Knowledge Based Systems. URL: http://pirata.ugr.es/ipmu96.html ***************************************************** * IPMU'96 * * * * Information Processing and Management of * * Uncertainty * * in Knowledge Based Systems * * * * Granada, Spain, July 1-5, 1996 * * * * URL: http://pirata.ugr.es/ipmu96.html * ***************************************************** Dear Colleague, Due to the increasing number of petitions we are receiving these days, the IPMU'96 Organization Committee has decided to postpone the deadlines for submitting. Enclosed you can find the new deadlines, as well as other useful information. ----------------------------------------------------------- INVITED SPEAKERS ----------------------------------------------------------- Prof. T. Terano (Japan). Kampe de Feriet Award Prof. A. F. Rocha (Brazil) Prof. H.J. Zimmermann (Germany) Prof. P. Bonissone (U.S.A.) Prof. R. Scozzafava (Italy) ----------------------------------------------------------- !!!! NEW DEADLINES !!!! ----------------------------------------------------------- December. 1 1995: Deadline for submission of papers. March. 1 1996: Notification of acceptance/rejection. April. 1 1996: Reception of final camera-ready. May 15 1996: Deadline for early registration. July 1-5 1996: CONFERENCE. ----------------------------------------------------------- INSTRUCTIONS ----------------------------------------------------------- There will be a six page (two columns, 10 pt) limit on the final versions of accepted papers. Papers will be carefully reviewed and authors will be notified on the acceptance/rejection by March 1st, 1996. Final camera-ready copies for publication will be required by April 1st, 1996. Authors can submit through two alternative ways: - Sending three copies of each full paper, by surface mail to: IPMU'96 Dpto. Ciencias de la Computacion e Inteligencia Artificial. E.T.S.I. Informatica. Avda. Andalucia, 38 Universidad de Granada. 18071 Granada. Spain. - Sending an electronic mail to: ipmu96-submissions@robinson.ugr.es or alternatively to ipmu96@robinson.ugr.es In this case, the following information must be included (in this order): a) Paper title (plain text) b) Author's names, including professional status. c) Surface mail and e-mail address for a contact author (plain text) d) A short abstract, including keywords or topic indicators (plain text) e) Paper body in postscript format (be sure it is not coded or compressed) ----------------------------------------------------------- TRAVELLING TO GRANADA ----------------------------------------------------------- Granada, a world-famous city, whose history spans over thousand years, also has outstanding features as a modern conference town. The Alhambra, the city's monuments, cultural and University traditions, as well as excellent leisure facilities, good restaurants, lively night life, the Sierra Nevada mountains and the Coast, all attract thousand of visitors to Granada every year. You can visit our city through URL: http://www.pirata.ugr.es/ipmu96.html For accommodation facilities, please contact with the travel agency: Viajes Bonal Avda. de la Constitucion 19 18014 - Granada. Spain Phone: + 34 58 276312 Fax: + 34 58 291967 ----------------------------------------------------------- PAYMENT ----------------------------------------------------------- Transfer to CONGRESO INTERNACIONAL IPMU96. Bank: CAJA GENERAL DE AHORROS DE GRANADA Account: 2031.0234.81.0100058503 Address: Avda. Andalucia s/n. Edificio Samoa 18014 Granada. Spain ----------------------------------------------------------- OTHER USEFUL INFORMATION ----------------------------------------------------------- The frequency of International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems will be increased from the current 4 issues per year to 6 issues, starting from 1996. In recognition of the association between the IPMU community and the journal, the editors of World Scientific Publishing are pleased to inform all the members of the IPMU community, that there is a special subscription rate of US$60 per year for the journal. This special rate remains the same as previous years, despite the increase in frequency. So then, subscribe the journal today! All you need to do is to write/fax in to: World Scientific Publishing Co. Pte. Ltd. Block 1022 Tai Seng Avenue #05-3520 Tai Seng Industrial Estate Singapore 538890 Republic of Singapore Tel: 65-3825663, Fax: 65-3825919 or simply send an email message to any one of the following addresses: worldscp@singnet.com.sg (Singapore office) wsped@singnet.com.sg (Editorial Dept, Singapore) wspmkt@singnet.com.sg (Marketing Dept., Singapore) In any case, please include the acceptance notification of your paper to IPMU'96, or the payment receipt if you are attending to the conference. >~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~