*
C. Mena, Mining The Net ? *
B. Masand, 'Neural' Computer Program Digs Up Metal Deposits Publications: *
GPS, Wanted: Short Application Descriptions for DMKD journal *
E. Rigdon, Marketing News on Data Mining and War Stories Meetings: *
J. Han, Last CFP: SIGMOD'97 Data Mining Workshop,
*
I. Haimowitz, CFP: AAAI-97 workshop on AI Approaches to
Fraud Detection and Risk Management
--
Discovery in Databases (KDD) community, focusing on the latest research and
applications.
Submissions are most welcome and should be emailed,
with a DESCRIPTIVE subject line (and a URL, when available) to kdd@gte.com
To subscribe, email to kdd-request@gte.com
message with
subscribe kdd-nuggets
in the first line (the rest of the message and subject are ignored).
See
Nuggets frequency is approximately 3 times a month.
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
********************* 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 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
It isn't that they can't see the solution. It's that they can't see the
problem. -G.K. Chesterton Previous1NextTop
Date: Fri, 10 Jan 1997 10:43:36 -0500
From: Phil Margolis (pmarg@sandybay.com)
To: kdd@gte.com
Subject: Your page selected by PC Webopaedia
Dear Webmaster:
Congratulations. The following URL at your site has been selected to be
included in the PC Webopaedia:
[P.S. please cc to kdd@gte.com
-- I think many Nuggets readers would also
be interested. GPS]
Previous3NextTop
Date: Thu, 16 Jan 1997 12:18:05 -0500
From: gps@gte.com
(Gregory Piatetsky-Shapiro)
Subject: Short Application Descriptions for DMKD journal
For the Data Mining and Knowledge Discovery journal,
we are looking for short application summaries (2-4 pages) describing
significant and successful deployed applications.
Publishing in Data Mining and Knowledge Discovery journal
would give an application a wide exposure to a community of data mining
professionals. The review and turnaround on those papers is very
quick. For more details, please email me at gps@gte.com
.
Gregory Piatetsky-Shapiro, Editor
Data Mining and Knowledge Discovery journal
Previous4NextTop
Date: Tue, 14 Jan 1997 14:10:01 -0500
From: ED RIGDON (MKTEER@langate.gsu.edu)
Subject: DM article in Marketing News and War Stories ?
Marketing News, the biweekly magazine of the American Marketing
Association, included my short article on data mining in a 'Special
Focus' section on marketing research (January 6, 1997, p. 8). The
article is titled (by the editor), 'Data Mining Gains New
Respectability.' While stripping all Web addresses from the article,
the editor added a 'Marketing Lesson' box with three highlighted
points:
'Data mining represents a promising approach to maximizing the
value of customer information.
A fully realized system adapts to the user so that its
performance (as perceived by the user) improves over time.
Marketers are moving to build the massive data bases that will
make data mining possible.'
I am also working on a KDD/data mining article for Marketing
Management magazine, an AMA quarterly aimed at senior marketing
managers, and I could use some help. The article aims to provide a
practical perspective on data mining system development. In this
regard, I would appreciate mail (erigdon@gsu.edu)
from both clients
and suppliers regarding their experiences. I would like to hear your
views on:
1. what went right, and what went wrong?
2. what elements of your situation were influential in producing these results?
3. in concrete terms, what payoffs has your KDD effort produced, and at what cost?
Yes, I'm looking for 'war stories.' And because the aim of the
article is practical advice, I would appreciate mailing from suppliers
that describe their products--e-mail to erigdon@gsu.edu,
fax (404)
651-4198, s-mail to Edward Rigdon, Marketing Department, Georgia State
University, University Plaza, Atlanta, GA 30303.
Thanks--
Ed Rigdon
erigdon@gsu.edu
Previous5NextTop
Date: Tue, 14 Jan 1997 14:08:38 -0500
From: brij@gte.com
(Brij Masand)
Subject: 'Neural' Computer Program Digs Up Metal Deposits
[Copyright, Reuters Ltd. All rights reserved ]
'Neural' Computer Program Digs Up Metal Deposits
LONDON - A self-teaching computer program has been launched which
trawls through reams of data to spot metal deposits in days where it
would take a team of mining geologists months, its developers says.
Britain's Neural Technologies, a privately owned company, said
Prospect Explorer has attracted an enthusiastic response from experts
who have practiced with the new program.
'There has only been massive enthusiasm,' said Peter Baxendale,
commercial director of Neural Technologies, which developed Prospect
Explorer with Australian copper and gold mine Straits
Resources. 'Nobody has seen anything like it.'
The patented program, which has been developed over the past three
years, can run on a simple personal computer and will cost about
40,000 pounds for an annual license.
A more sophisticated and faster version will be available in the
summer at 200,000 to 250,000 pounds per license.
Prospect Explorer, whose 'neural technology' means it learns from
experience like the brain, processes all kinds of data, such as soil
chemistry, radioactivity and magnetism, to detect geological patterns
and anomalies overnight.
Further analysis of the anomalies may reveal within a week whether an area has mining potential.
To test Prospect Explorer, Straits Resources used it to investigate a
100 square-kilometer plot where it was already active.
Within six days the program found Straits Resources' Girlambone copper
mine and several other potential sites.
It had taken a team of up to eight Straits Resources geologists using
conventional data analysis two months to complete their own
investigation of the region.
At a presentation in London, Neural Technologies said mines can reduce
the time from selecting land to actual drilling by about 25 times if
they use Prospect Explorer.
'This saves time...and mines will have to do much less test drilling,
which is very expensive,' Baxendale told journalists.
Prospect Explorer will be marketed by a joint venture recently set up
by Neural Technology and Straits Resources.
Straits will have 70 percent of the Neural Mining Solutions venture,
while Neural Technologies holds the rest.
Neural Mining Solutions is already in talks with 'several leading
mining houses' to sell Prospect Explorer, Baxendale said.
Copyright, Reuters Ltd. All rights reserved
Previous6NextTop
From: Jiawei Han (han@cs.sfu.ca)
Date: Thu, 9 Jan 1997 13:54:51 -0800 (PST)
Subject: Last Call for Papers: SIGMOD'97 Data Mining Workshop
============================================================================
Call For Papers
============================================================================
Workshop on Research Issues on Data Mining and Knowledge Discovery (DMKD'97)
in cooperation with ACM-SIGMOD'97
Tucson, Arizona, May 11 (Tentative) 1997
==========================================
(URL:
Mining knowledge from large databases and data warehouses is a promising
research area, with high application potential due to the huge amounts of
data accumulated in databases, data warehouses, and other information
repositories. Data mining has attracted people from many different fields,
including database systems, data warehouses, machine learning, knowledge
acquisition, statistics, information retrieval, and data visualization.
In June 1996, we organized a SIGMOD workshop on research issues on data
mining and knowledge discovery. It was well attended and was widely
considered to be successful in creating a forum for database researchers
to exchange their research ideas and results in data mining. To continue
to provide such a forum, we are organizing the second workshop in cooperation
with SIGMOD.
FORMAT
The workshop will be held one day before the SIGMOD/PODS'97 conference.
The plan is to have a full-day workshop, consisting of invited talks,
paper presentation/discussion sessions, a system demo session, and a
panel discussion session. If there are a good number of submissions
and enough interest, we may organize a poster session in parallel with
a data mining system demonstration session.
TOPICS
Major topics of interest include but are not limited to:
Foundations/principles of data mining
Data mining methods and algorithms
Association, classification, and prediction
Concept description: characterization and discrimination
Trend/deviation analysis and outlier detection
Integration of data mining and data warehousing
Mining knowledge in multidimensional databases
Integration of deduction, induction, and OLAP
Statistics, probability and uncertainty in data mining
Interestingness of discovered patterns
Efficiency and scalability in data mining
Parallel and distributed mining algorithms
Languages and interfaces for data mining
Visual data mining and visualization in data mining
Data mining systems and implementations
Data mining toolkits and methodologies
Performance and benchmarks of data mining systems
Mining spatial, temporal, and multimedia data
Data mining in heterogeneous databases and WWW
Integrated discovery systems
Successful data mining application examples
New application challenges and requirements
Inadequacy of current data mining mechanisms
Security and social impact of data mining
Influence of data mining to the advances of database systems
SUBMISSION AND REVIEWS OF POSITION PAPERS and RESEARCH PAPERS.
Authors are invited to submit position papers (limited to 5 pages) and/or
short research papers (or extended abstracts) (limited to 10 pages) on the
above topics. WE ENCOURAGE ELECTRONIC SUBMISSIONS IN THE FORM OF POSTSCRIPT,
LATEX, ETC. but limited to the std 8.5x11 sized paper. If hardcopies are
submitted, five copies will be required. Each submitted paper will be
reviewed by at least three program committee members. Selected papers
from this workshop will be considered for a special issue of the journal:
'Data Mining and Knowledge Discovery'.
PROGRAM COMMITTEE
Rakesh Agrawal, IBM Almaden Research Center, USA
Inderpal Bhandari, IBM T.J. Watson Research Center, USA
Nick Cercone, University of Regina, Canada
Ming-Syan Chen, National Taiwan University, Taiwan
David W. Cheung, University of Hong Kong, Hong Kong
Umeshwar Dayal, Hewlett-Packard Laboratories, USA
Usama M. Fayyad, Microsoft Research, USA
Brian Gaines, University of Calgary, Canada
Randy Goebel, University of Alberta, Canada
Jiawei Han, Simon Fraser University, Canada
Tomasz Imielinski, Rutger University, USA
Bala Iyer, IBM Database Technology Institute, USA
Daniel A. Keim, University of Munich, Germany
Willi Kloesgen, GMD, Germany
Hans-Peter Kriegel, University of Munich, Germany
Laks V.S. Lakshmanan, Concordia University, Canada
Hongjun Lu, National University of Singapore, Singapore
Heikki Mannila, University of Helsinki, Finland
Shinichi Morishita, IBM Tokyo Research Center, Japan
Shamkant B. Navathe, Georgia Institute of Technology, USA
Raymond Ng, University of British Columbia, Canada
Shojiro Nishio, Osaka University, Japan
Gregory Piatetsky-Shapiro, GTE Laboratories, USA
Wei-Min Shen, University of Southern California, USA
Ramakrishnan Srikant, IBM Almaden Research Center, USA
Shalom Tsur, Hitachi America Ltd., USA
Alexander Tuzhilin, New York University, USA
Jeffrey D. Ullman, Stanford University, USA
Philip S. Yu, IBM T.J. Watson Research Center, USA
Carlo Zaniolo, Univ. of California at Los Angeles, USA
ORGANIZING COMMITTEE
Jiawei Han, Simon Fraser University, Canada
Laks V.S. Lakshmanan, Concordia University, Canada
Raymond Ng, University of British Columbia, Canada
IMPORTANT DATES
Submissions Due: January 24, 1997
Acceptance Notice: March 14, 1997
Final Version due: April 11, 1997
Send a short abstract of at most 150 words in ascii to rng@cs.ubc.ca
by
January 24, 1997.
Five hard copies or one electronic copy of the paper should be submitted
by January 24, 1997, to
Dr. Raymond Ng
Department of Computer Science
University of British Columbia
Vancouver, B.C., V6T 1Z4, Canada
rng@cs.ubc.ca
Previous7NextTop
From: HOUSTON@KGNVMC.VNET.IBM.COM
Date: Fri, 10 Jan 97 03:00:28 EST
Subject: ASEAN Data Mining/ Data Warehousing Seminar
Content-Length: 991
AGENDA...
-A Customer Perspective: DW at Bell Canada
by Bell Sygma Telecom
-Intelligent Data Mining with IBM
-DW with ORACLE and S/390
-DW with SnapShot - Create a virtual database
fast with SnapShot
-Sizzling Queries with DB2
LOCATION and DATES...
Philippines 18 February The Penisula Manila RSVP 819-2426
(Garcia, Villa, Balagtas, Level 2)
Malaysia 20 February Kuala Lumpur Hilton RSVP (03) 717-7890
(Windows on KL 1, Level 30)
Singapore 21 February Marina Mandarin RSVP 1800-320-1234
(Orion Room, Level 2)
Thailand 25 February Amari Watergate RSVP 273-4444
(Amari Watergate Ballroom C, Level 6)
Indonesia 27 February Hotel Sari Pan Pacific RSVP 251-2922
(Istana Ballroom) Previous8NextTop
Subject: IDA-97 REMINDER
Date: Mon, 13 Jan 1997 16:36:24 +0000
From: Michael Berthold (berthold@ira.uka.de)
IDA-97 REMINDER
The Second International Symposium on Intelligent Data Analysis (IDA-97)
is to be held in Birkbeck College, University of London, 4th-6th August
1997. The deadline for submissions is February 1st, 1997. The details
regarding IDA-97 can be found at
IDA-97 Administrator
Department of Computer Science
Birkbeck College
Malet Street
London WC1E 7HX, UK
E-mail: ida97-enquiry@dcs.bbk.ac.uk
Tel: (+44) 171 631 6722
Fax: (+44) 171 631 6727
Previous9NextTop
From: Matthias Klusch (mkl@informatik.uni-kiel.de)
Subject: CIA-97 Workshop on Cooperative Information Agents
Date: Tue, 14 Jan 1997 12:24:39 MEZ
*****************************************************************************
First International Workshop CIA-97 on
COOPERATIVE INFORMATION AGENTS
26th (Wed) - 28th (Fri) of February 1997
University of Kiel, Computer Science Department,
Kiel, Germany
*****************************************************************************
The workshop CIA-97 will be held in cooperation with the research groups on
- Distributed Artificial Intelligence (DAI),
- Database Systems, and
- Methods for Information Systems Development (EMISA)
of the German Society for Computer Science (GI).
...........................................................................
Invited Speakers
Mike Wooldridge (Mitsubishi Electric Digital Library Group, UK)
Misbah Deen (University of Keele, UK)
Sonia Bergamaschi (University of Modena, Italy)
Hans-Dieter Burkhard (Humboldt University of Berlin, Germany)
Larry Kerschberg (George Mason University, Fairfax, USA)
Gottfried Vossen (University of Muenster, Germany)
Aris Ouksel (University of Illinois at Chicago, USA)
...........................................................................
Previous10NextTop
Date: Tue, 14 Jan 1997 12:36:17 -0500
From: haimowit@dogwood.crd.ge.com
(Ira Haimowitz)
To: kdd@gte.com
Subject: AAAI-97 workshop in fraud/risk
Dear Gregory:
Below is a call for papers for a 1-day workshop
on artificial intelligence for fraud detection and risk management.
Data mining related entries are most welcome.
Thanks very much,
Ira Haimowitz
GE Corporate R & D
---------------------------------
CALL FOR PARTICIPATION
The AAAI-97 Workshop on
AI Approaches to Fraud Detection and Risk Management
DESCRIPTION:
Fraud detection and risk management involve monitoring the behavior of
populations of users in order to estimate, detect or avoid undesirable
behavior. Undesirable behavior is a broad term including delinquency,
fraud, intrusion and account defaulting. This workshop will bring together
researchers in these areas to discuss approaches and experiences in dealing
with the critical issues:
- large volumes of data
- highly skewed distributions