KDD Nuggets 95:31, e-mailed 95-12-03 Contents: News: * GPS, Langley and Simon in Nov 95 Comm of ACM on Applications of ML * IDIS, Information Discovery, Inc Home page http://www.datamining.com/ Publications: * GPS, IBM white paper on Data Mining http://booksrv2.raleigh.ibm.com/cgi-bin/bookmgr/bookmgr.cmd/BOOKS/datamine/ Siftware: * J. Betak, OC1 and PEBLS decision tree software for DOS, ftp://ftp.gmd.de/ml-archive/software.html Meetings: * L. Huan, Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD97), Singapore, February 1997, http://www.iscs.nus.sg/conferences/pakdd97.html * J. Wnek, CFP: Multistrategy Learning Workshop MSL'96, May 23-25, 1996, West Virginia, http://www.mli.gmu.edu/msl96.html -- The KDD Nuggets is a moderated mailing list focusing on Data Mining and Knowledge Discovery in Databases (KDD) research and development. Contributions are welcome and should be emailed, with a DESCRIPTIVE subject line (and a URL, when available) to . E-mail add/delete requests to . Nuggets frequency is approximately weekly. Back issues of Nuggets, a catalog of S*i*ftware (data mining tools), and a wealth of other information on Data Mining and Knowledge Discovery is available at Knowledge Discovery Mine site, URL . -- 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 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Optimists and pessimists are equally accurate, but optimists live longer. Observation by an optimist >~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Date: Wed, 22 Nov 1995 16:13:29 -0500 From: gps0 (Gregory Piatetsky-Shapiro) Subject: Langley and Simon in Nov 95 Comm of ACM on Applications of ML Nov 95 Communications of ACM is a special issue on New Horizons in Commercial and Industrial AI. In particular, it has an interesting article by Pat Langley and Herbert Simon on Applications of Machine Learning and Rule Induction, which includes a number of applications relevant to Data Mining and Knowledge Discovery. These include Making Credit Decisions at American Express UK by deriving rules from historical data, developed by Michie et al. Deriving new rules for Diagnosis of Mechanical Devices, developed by Giordana, Neri and Saitta in Italy. Automatic Classification of Celestial Objects, developed by Fayyad, Smyth, Weir, and Djorkovski at JPL. (Their system, SKICAT, was widely reported at KDD meetings) and numerous other examples. Langley and Simon also give some guidance for key steps in developing practical ML systems: formulating the problem determining the representation Collecting the training data Evaluating the learned knowledge Fielding the knowledge base They suggest that the main source of power comes not necessarily from the best induction method -- in fact, most of the application efforts have used well understood methods rather than the latest research results. Instead, most of the power comes from the proper formulation of the problems and from crafting the representation to make the learning tractable. >~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Date: Mon, 27 Nov 95 18:02:15 -0500 From: Gregory Piatetsky-Shapiro Subject: Information Discovery, Inc. Home Page Information Discovery, Inc announced its home page http://www.datamining.com/ (which is mapped into http://datamine.inter.net/datamine/) The home page contains info on the company, and also on * What is Data Mining? * Information Discovery Products and Services * Download a Demo Disk The homepage http://www.datamining.com/ is mapped by DNS into http://datamine.inter.net/datamine/ >~~~Publications:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Date: Sun, 3 Dec 1995 16:13:37 -0500 From: gps0 (Gregory Piatetsky-Shapiro) Subject: IBM White Paper on Data Mining IBM has published a white paper on Data Mining. It is available at http://booksrv2.raleigh.ibm.com/cgi-bin/bookmgr/bookmgr.cmd/BOOKS/datamine/ Abstract: Competitive business pressures and a desire to leverage existing information technology investments have led many firms to explore the benefits of data mining technology. This technology is designed to help businesses discover hidden patterns in their data -- patterns that can help them understand the purchasing behavior of their key customers, detect likely credit card or insurance claim fraud, predict probable changes in financial markets, etc. This paper explores data mining, its potential benefits to users, and IBM's activities in this area. It also explains how data mining activities can be integrated within an existing user environment, including those that already make use of data warehousing. >~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ >~~~Positions:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ >~~~Siftware:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Return-Path: From: jbetak@rz.fh-augsburg.de (Betak Juraj) Subject: OC1 and PEBLS for DOS Date: Sat, 2 Dec 1995 15:36:46 +0100 (MET) Dear Gregory, OC1 and PEBLS, both developed at the Johns Hopkins University, and kindly made avaible to the public, have been successfully ported to DOS and can be accessed at GMD' ML-Archive. Interested people looking for PC software might want to point their WWW browser at: ftp://ftp.gmd.de/ml-archive/software.html And have a look under "OC1" or "PEBLS", where they find the PC port and a link to the originating source at JHU. Many thanks to Sreerama Murthy, Steven Salzberg and Stefan Wrobel for their kind support. Best regards Juraj P.S. A short note for folks, who are not familiar with the software mentioned. OC1 is decision tree oriented, PEBLS implements a nearest neighbour algorithm. Both allow a very liberal control and include some other approaches for comparison purposes. For more information, please email murthy@scr.siemens.com (OC1 - primary contact) salzberg@cs.jhu.edu (PEBLS/OC1) >~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ >~~~Meetings:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ From: Liu Huan Subject: Annoucement of PAKDD97 (new format) To: liuh@iscs.nus.sg (Liu Huan) Date: Thu, 23 Nov 1995 11:09:34 +0800 (GMT-8) FIRST PACIFIC-ASIA CONFERENCE on KNOWLEDGE DISCOVERY and DATA MINING (PAKDD97) Singapore, February 1997 (Co-located with 2nd Pacific-Asia Conference on Expert Systems/ 3rd Singapore International Conference on Intelligent Systems) ------------------------------------------------------------------------------- The first Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD97) will be held in Singapore in February, 1997. As the range of computer applications is broadening, more and more computer generated data are produced. In order to overcome the situation of ``data rich and knowledge poor'', KDD is becoming the focus of many fields from Intelligent Databases, Machine Learning to Statistics. The aims of the conference are to cover all aspects of KDD, to bring together researchers and practitioners from basic and applied research and information industries, and to push forward the state-of-art of KDD. The conference technical programme will include paper presentations, posters, invited talks, panels, and tutorials in a two-day event. The following topics of interests serve as an indication and include, but not limited to: Knowledge Representation and Acquisition in KDD Data Mining and Data Warehousing Data Cleaning, Preprocessing and Postprocessing Data and Dimensionality Reduction Knowledge Reuse and Role of Domain Knowledge Data Mining Tools KDD Framework and Process Security and Privacy Issues in KDD Mining in-the-Large vs Mining in-the-Small Management Issues in KDD Machine Learning, Statistical and Visualization Aspects of KDD Successful/Innovative Applications in Science, Government, Business and Industry The proceedings will be published by an international publisher and will be available at the conference. Conference Chair: Hing-Yan Lee Japan-Singapore AI Centre Information Technology Institute Programme Co-Chairs: Hongjun Lu Hiroshi Motoda Dept of Info. Sys. & Comp. Sci. Institute of Sci. & Indus. Research National Univ of Singapore Osaka University, Japan Publicity Chair: Conference Secretary: Huan Liu, DISCS, NUS Hwee-Leng Ong, JSAIC, ITI (Program Committee to be decided) WWW URL: http://www.iscs.nus.sg/conferences/pakdd97.html PAKDD97 is organized by Information Technology Institute and National University of Singapore ------------------------------CUT FROM HERE------------------------------------ Please keep me on your mailing list. I am interested in attending/receiving information on PAKDD97. Name: E-mail: Tel: Fax: Full Postal Address: Please send me Call for Papers _ , registration form _ , I intend to submit a paper _ . Mail, e-mail or fax to Ms Hwee-Leng Ong, Japan-Singapore AI Centre, 11 Science Park Road, Singapore Science Park II, Singapore 117685, E-mail: hweeleng@iti.gov.sg, Fax: (+65) 779 1827. >~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Date: Mon, 27 Nov 95 15:49:19 EST From: jwnek@aic.gmu.edu (Janusz Wnek) Subject: MSL'96 Workshop (revised) CALL FOR PAPERS The Third International Workshop on Multistrategy Learning (MSL'96) May 23-25, 1996 Hilltop Inn, Harpers Ferry, West Virginia http://www.mli.gmu.edu/msl96.html ------------------------------------------------------------------------- The rapid expansion of machine learning methods, approaches and paradigms creates a strong need for investigating their interrelationships, and the development of methods for their synergistic integration. Multistrategy learning workshops (MSL) provide a forum for presenting and discussing research on these topics, and related issues, such as the cross-fertilization of machine learning and cognitive science research, learning with large knowledge bases, goal-oriented learning agents, and multistrategy approaches to knowledge discovery. Topics of interest include, but are not limited to, the following: - comparative studies of learning strategies, methods and paradigms - cognitive models of learning, inference and discovery - intelligent learning agents and complex adaptive systems - user-oriented learning in distributed information systems (e.g., WWW) - role of learning goals and learning in large knowledge systems - knowledge representation, acquisition and reuse in multistrategy learning and inference systems - advanced applications of multistrategy learning and knowledge discovery MSL '96 will be held in the picturesque and historic Harpers Ferry, located at the intersection of Virginia, West Virginia and Maryland. Harpers Ferry is easily accessible from Dulles airport and from the Washington, D.C. railway station. To maintain a true workshop atmosphere, the attendance will be limited to approximately 60 participants. The workshop is organized by Machine Learning and Inference Laboratory at George Mason University, and sponsored by National Science Foundation, and Office of Naval Research. IMPORTANT DATES: February 10, 1996: Paper submission deadline. Send four copies (no more than 15 single-spaced pages) to MSL '96, Machine Learning and Inference Laboratory Attention J. Wnek, MS 4A5, SITE 2 George Mason University 4400 University Drive, Fairfax VA 22030-4444, USA March 25, 1996: Author notification April 25, 1996: Dealine for final manuscript For more information see http://www.mli.gmu.edu/msl96.html INVITED SPEAKERS: John Anderson, Carnegie Mellon University Jaime Carbonell, Carnegie Mellon University Hugo de Garis, ATR, Kansai Science City Laveen Kanal, University of Maryland & LNK Inc. Doug Lenat, CycCorp Doug Medin, Northwestern University Marvin Minsky, Massachussets Institute of Technology Michael Pazzani, University of California at Irvine Luc de Raedt, Catholic University of Leuven Paul Rosenbloom, University of Southern California Lorenza Saitta, University of Torino Claude Sammut, University of New South Wales Derek Sleeman, University of Aberdeen PROGRAM COMMITTEE: John Anderson, Carnegie Mellon University Franscesco Bergadano, University of Messina Jaime Carbonell, Carnegie Mellon University Marie desJardins, SRI International Hugo de Garis, ATR, Kansai Science City Diana Gordon, Navy Research Laboratory Kenneth Haase, Massachusetts Institute of Technology Heedong Ko, Korea Institute of Technology Yves Kodratoff, University of Paris South Stan Matwin, University of Ottawa Doug Medin, Northwestern University Raymond Mooney, University of Texas at Austin Stephen Muggleton, Oxford University Michael Pazzani, University of California at Irvine Luc de Raedt, Catholic University of Leuven Ashwin Ram, Georgia Institute of Technology Lorenza Saitta, University of Torino Claude Sammut, University of New South Wales Jude Shavlik, University of Wisconsin Derek Sleeman, University of Aberdeen Gheorghe Tecuci, George Mason University and Romanian Academy ORGANIZERS: Ryszard S. Michalski, Chair, George Mason University and Polish Academy of Sciences (michalski@gmu.edu) Janusz Wnek, Co-Chair, George Mason University (jwnek@gmu.edu) LOCAL ARRANGEMENTS: Abhay Kasera, George Mason University (akasera@aic.gmu.edu; 703-993-1714; Fax 703-993-3729)