(text)
Ismail Parsa, KDD-98 Exhibit Presentations available on the web
Publications:
(text)
Gregory Piatetsky, DB2 Online Fall 1998 on Text and Data Mining
(text)
Douglas Fisher, Special Issue of MLJ on Unsupervised Learning
(text)
Gheorghe Tecuci, New Book 'BUILDING INTELLIGENT AGENTS:
An Apprenticeship Multistrategy Learning Theory, ... '
(text)
Vincent Corruble, CFP: IJHCS special issue on Machine Discovery
Tools/Services:
(text)
Sergei Ananyan, TextAnalyst - new text mining solution from Megaputer
(text)
Peter Raeth, Free Adaptive Automation Web Site Available
Positions:
(text)
Andreas Weigend, Faculty Openings at NYU/Stern Information Systems
(text)
Haym Hirsh, faculty recruitment in AI at Rutgers
(text)
Michael Pazzani, Graduate Programs at UCI in KDD and related areas
Meetings:
(text)
Trish Carbone, Federal Data Mining Symposium & Exposition '99
McLean, VA, March 9-10, 1999
(text)
Domenico Talia, CFP: EURO-PAR'99 - High-Performance DM and KDD
Toulouse, France, August 31 - September 3, 1999
(text)
IAT99, IAT'99: 1st Asia-Pacific Conference on Intelligent
Agent Technology, Hong Kong, December 15-17, 1999
--
Knowledge Discovery Nuggets (TM) or KDNuggets for short, is an
electronic newsletter focusing on the latest news, publications, tools,
meetings, and other relevant items in the Data Mining and Knowledge Discovery
field. KDNuggets is currently reaching over 6000 readers in 75+ countries
2-3 times a month.
Items relevant to data mining and knowledge discovery are welcome
and should be emailed to gps
in ASCII text or HTML format.
An item should have a subject line which clearly describes
what is it about to KDNuggets readers.
Please keep calls for papers and meeting announcements
short (50 lines or less of up to 80-characters), and provide a web site for
details, such as papers submission guidelines.
All items may be edited for size.
Back issues of KD Nuggets, a catalog of data mining tools
('Siftware'), pointers to data mining companies, relevant websites,
meetings, etc are available at KDNuggets Directory at http://www.kdnuggets.com/
********************* Official disclaimer ***************************
All opinions expressed herein are those of the contributors and not
necessarily of their respective employers (or of KD Nuggets)
*********************************************************************
~~~~~~~~~~~~ Quotable Quote ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Only in America...do we buy hot dogs in packages of ten and buns in
packages of eight...
thanks to Tony Hu. Previous1NextTop
Date: Thu, 19 Nov 1998 14:20:17 -0500
From: Ismail Parsa, iparsa@epsilon.com
Subject: KDD-98 Exhibit Presentations Web Site
Web: http://www.epsilon.com/new
A web site collecting the KDD-98 Exhibits presentations is now
available on the web at http://www.epsilon.com/new.
The following three exhibit presentations were given at the 4th
International Conference on Knowledge Discovery and Data Mining
(KDD-98):
Data Mining in the Real World
-----------------------------
Gordon Linoff, co-author of Data Mining Techniques
(Data Miners, gordon@data-miners.com)
Friday, August 28, 1998 (4:30-5:15PM)
This presentation discusses the issues of data mining in the real
world. It touches on the relationship of data mining with a data
warehouse (is it really easier?) and on issues related to managing
data and choosing particular techniques.
Data Mining On the Internet
---------------------------
'Overview, Algorithmic Challenges and Applications.'
Shivakumar Vaithaynathan (IBM Research, vaithyan@us.ibm.com)
Saturday, August 29, 1998 (12:00-12:45PM)
The advent of the World Wide Web has caused a dramatic increase
in the usage of the Internet. The resulting growth in on-line
information combined with the almost chaotic nature of the web
necessitates the development of powerful yet computationally
efficient algorithms. This presentation provides examples of
applications where data mining could be applied and then focuses
on the algorithmic challenges that go along. New algorithms and
results are provided.
Data Mining Tools
-----------------
Ismail Parsa (Epsilon, iparsa@epsilon.com)
Saturday, August 29, 1998 (4:30-5:15PM)
The data mining tools marketplace is diverse. There are tools
that offer broad-based data mining capability, tools aimed at
solving the problems of a particular industry, tools combined
with a service offering, black-box tools and vendors/tools offering
custom solutions such as CRM, campaign management, etc. This
presentation segments the data mining tools marketplace then shows
how to differentiate between the many data mining tools for the
best return on investment. Summary results of a real-life data mining
tool evaluation case study are explored and explained.
Ismail Parsa
Epsilon
50 Cambridge Street
Burlington MA 01803 USA
Previous2NextTop
Date: Sun, 22 Nov 1998
From: Gregory Piatetsky-Shapiro gps
Subject: DB2 Online Fall 1998 on Text and Data Mining
DB2 Online Fall 1998 Magazine features two interesting articles on
Text Mining: Beyond Search Technology, by Patricia Soto
How to implement intelligent search capabilities in your
organization using text mining technology.
Mining Customer Data, by Gary Saarenvirta
A step-by-step look at a powerful data mining methodology for
evaluating customer value: customer clustering and segmentation.
Call for Papers
Special Issue of Machine Learning on Unsupervised Learning
Doug Fisher
Special Issue Editor
Several forms of unsupervised learning extract
relationships from data that can be then exploited for
inference. The primary unsupervised techniques include
clustering, learning (usually Bayesian) belief networks,
and learning association rules. The unsupervised 'pattern'
or 'concept' learning methods that are of most interest in
this special issue differ from supervised concept learning
methods in that there is no single, dependent variable,
dimension, or predicate that is the *a priori* focus of
inference. Rather, an unsupervised method may support
inference along more than one dimension (variable, property),
typically many dimensions/properties.
Authors are encouraged to submit papers in the primary
unsupervised learning paradigms of clustering, belief-
network learning, and association-rule learning (and
possibly others) for consideration as contributions to
the Special Issue on Unsupervised Learning of the
journal, Machine Learning. Articles that relate
different paradigms are especially welcome.
Previous4NextTop
Date: Tue, 10 Nov 1998 17:28:15 -0500
From: Gheorghe Tecuci tecuci@gmu.edu
Subject: New Book 'BUILDING INTELLIGENT AGENTS: An Apprenticeship Multistrategy
Learning Theory, Methodology, Tool and Case Studies'
This book presents a theory, methodology and tool for building
intelligent agents, along with detailed case studies. The most
significant, and unique, characteristic of building these agents is that
a person directly teaches them how to perform domain-specific tasks in
much the same way he or she would teach a student or apprentice: by
giving the agent examples and explanations, and by supervising and
correcting its behavior. This approach, in which the agent learns its
behavior from its teacher, integrates many machine learning and
knowledge acquisition techniques, taking advantage of their
complementary strengths to compensate for each other weaknesses. As a
consequence, it significantly reduces the involvement of a knowledge
engineer in the process of building an intelligent agent. The book is
unique in the comprehensive coverage of its subject. The first part of
the book presents an original theory for building intelligent agents and
a methodology and tool that implement the theory. The second part of the
book presents complex and detail case studies of building different
types of agents: an educational assessment agent that enhances the
capability, generality and usefulness of an educational system for
teaching higher-order thinking skills in the context of history; a
statistical analysis assessment and support agent to support a
university-level introductory science course; an engineering design
assistant that cooperates with its user in configuring computer systems;
and a virtual military commander integrated into a distributed
interactice simulation environment.
CONTENTS:
Preface. Intelligent Agents. General Presentation of the Disciple
Approach for Building Intelligent Agents. Knowledge Representation and
Reasoning. Knowledge Acquisition and Learning. The Disciple Shell and
Methodology. Case Study: Assessment Agent for Higher-Order Thinking
Skills in History. Case Study: The Statistical Analysis Assessment and
Support Agent. Case Study: Design Assistant for Configuring Computer
Systems. Case Study: Virtual Agent for Distributed Interactive
Simulations. Selected Bibliography of Machine Learning, Knowledge
Acquisition, and Intelligent Agents Research. Notation. Subject Index.
International Journal of Human-Computer Studies
Special Issue on Machine Discovery
Call for Papers
Focus of the Special Issue
Scientific discovery is a human and social process that has
attracted attention from a growing portion of the AI community,
as well as from neighbouring disciplines such as philosophy and
psychology. It is an important area for the study of creativity,
itself a fundamental subject for Artificial Intelligence research.
A number of recent meetings have been dedicated to scientific discovery,
such as a AAAI Spring Symposium at Stanford University in 1995 and
a workshop at the European Conference on Artificial Intelligence in
Brighton (UK) in August 1998. Scientific Discovery was also the topic
of a special issue of the Artificial Intelligence Journal in 1997.
Given that the tasks tackled are usually complex, and require background
knowledge at the hypothesis generation and evaluation stages, most
systems involve an expert in the loop. The experts' primary role is
to evaluate hypotheses and the proposed new knowledge. Thus, for the
present, most scientific discovery systems are Cooperative, making this
a very suitable topic for a special issue of this journal.
Contribution are invited in the following areas:
(Cooperative) systems and tools to automate/aid scientific discovery
New computational models of scientific activity
Reports on new scientific findings resulting from the use of
computational tools performing non-trivial, high-end tasks
Lessons learned from earlier science (recent or not so recent) through
computational simulations and case studies
Psychological studies of the discovery process.
Megaputer Intelligence Inc. and MicroSystems Ltd. unveil TextAnalyst - a
unique intelligent tool for automated semantic analysis, summarization,
and navigation of texts written in natural language. The new solution
received the highest grades from all pilot users of the system during
the initial offering period. TextAnalyst performs a broad range of text
mining tasks completely automatically:
* Semantic network development for a corpus of analyzed texts;
* Text abstracting;
* Knowledge base creation - summarization and navigation through a set
of texts;
* Semantic search for information (with the creation of a sub-tree of
concepts related to the concept mentioned in the query);
* Hierarchical topic structuring;
* Text tagging - marking important concepts in the body of the text.
TextAnalyst utilizes an innovative neural network technology for a
homogeneous processing of texts from arbitrary fields. In combination
with other implemented advanced text processing techniques this approach
provides for the most efficient development of the semantic network of a
text. The developed semantic network is used by TextAnalyst for a
further multi-faceted analysis of texts.
Existing users of TextAnalyst include government offices, consulting and
law firms, medical centers, scientific organizations, electronic book
publishers, customer support centers, political institutions, and even
college students.
TextAnalyst effortlessly permits us to undertake preliminary data
reduction and analysis simultaneously without missing important
information. TextAnalyst is able to efficiently handle numerous and
often large (90+ pages apiece) text files without any problem.
Furthermore, the program is extremely user-friendly, says Eleanor
McLellan, Data Analyst at a Large Government Medical Research facility.
TextAnalyst facilitates a drastic reduction in the time required by an
analyst to grasp the meaning of a set of texts in any subject. In
addition, TextAnalyst furnishes many hints for a deeper comprehension of
the whole text, as well as its separate fragments, adds Peter
Makogonov, Ph.D., Deputy Director, Administration of Mayor, Moscow.
By introducing TextAnalyst Megaputer further expands its broad offering
of comprehensive data analysis solutions and becomes a key player in one
of the most dynamic segments of the information analysis market, notes
Sergei Arseniev, CEO of Megaputer. Now large corporate customers with
diverse data and text analysis needs can obtain a complete suite of best
analytical solutions from a single vendor, making the integration of
different applications simple. Megaputer's PolyAnalyst solution for data
mining and TextAnalyst solution for advanced text analysis go
hand-in-hand to help users derive more value from information.�
Megaputer is running a limited time introductory promotion for
TextAnalyst. A FREE evaluation copy of the program is available for
downloading from http://www.megaputer.com
Platforms: Windows 95/ 98/ NT
Pricing: The limited time promotional price for TextAnalyst is $283.
An additional internet purchase discount is 30% until January 1, 1999.
Adaptive Automation Resources is a free web site that categorizes
adaptive automation links. This site addresses the following major
areas: algorithms, statistics, ops research, graph analysis, expert
systems, fuzzy logic, neural networks, and evolutionary computation. It
organizes links to information that a broad audience should find
understandable and useful within the problem solving technology
continuum of advanced heuristic methods. Here is a place to find FAQs,
Newsgroups, Software, Books, Electronic Journals, and Hot Lists.
Adaptive automation is an exciting technology filled with opportunity to
solve seemingly intractable problems. It is also a powerful tool for
developing models of processes and systems that do not yield to
traditional analysis or constructive techniques. This site is available
for your use at: http://www.geocities.com/siliconvalley/lakes/6007.
As
always, your comments are welcome.
The Department of Information Systems at the Stern School of Business
at New York University has several faculty openings:
*** Faculty Positions
Applications are solicited for tenure-track positions at all levels
for the 1999-2000 academic year. Entry-level candidates must receive
the Ph.D. by summer 1999. Candidates must present evidence of
outstanding research and teaching performance in the application of
information technology to the solution of business problems. We are
especially interested in fields such as data driven learning and
knowledge discovery, the economics of information, electronic
commerce, organization design and change, and human-computer
interaction.
*** Visiting Faculty
Applications are solicited for full-time visiting positions at all
levels. Visiting faculty teach at the MBA or undergraduate level and
are active in the research activities of the department. Candidates
must present evidence of strong teaching and research performance.
NYU encourages applications from women and members of minority groups.
Please send your application material as soon as possible to:
Professor Jon Turner
Chair of the Recruiting Committee
Department of Information Systems
Leonard N. Stern School of Business
New York University
44 West 4th Street, K-MEC 9-72
New York, NY 10012-1126, USA
To speed up the process, you can also send the usual application
material to isnyu99@stern.nyu.edu,
or fax it to +1 212-995-4228.
If you are specifically interested in areas such as computational
finance, financial engineering, statistical artificial intelligence,
machine learning, neural networks, Bayes networks, graphical models,
computational intelligence, reasoning under uncertainty, data mining,
knowledge discovery, forecasting, time series prediction, etc., you
can also contact me directly through e-mail (aweigend@stern.nyu.edu).
Andreas S. Weigend
Associate Professor
Information Systems Department
Stern School of Business, NYU
44 W 4th St., K-MEC 9-74
New York, NY 10012-1126, USA
The Computer Science Department at Rutgers University is seeking
outstanding candidates for several tenure-track positions for the 1999
academic year. We are particularly interested in artificial intelligence,
bio-computing, cryptography, databases, digital libraries, distributed and
parallel systems, networking, operating systems, security, software
engineering, and theoretical computer science, although exceptionally
strong candidates in all areas are encouraged to apply.
The new Rutgers University strategic plan places computer science as an
area with one of the highest expected growth rates within the university.
Our department's $4 million of outside funding is distributed among
forty-five research projects, spanning the spectrum of computer science.
We are partners in major centers in theoretical computer science (DIMACS),
cognitive science (RuCCS), and wireless and mobile computing (WINLAB),
among others. (For more details, see http://www.cs.rutgers.edu.
Our
department also has close ties to industry including AT&T, Bellcore,
Hewlett-Packard, Lucent, NEC, Siemens, and Sun, as well as several
emerging companies in the New York area.
To apply please submit a curriculum vitae including names of at least four
professional references to:
Professor Eric Allender, Hiring Chair
Department of Computer Science
Rutgers, the State University
110 Frelinghuysen Road
Piscataway, NJ 08854-8019
by February 1, 1999, or send e-mail to hiring@cs.rutgers.edu
for further
information. We especially encourage applications from women and other
under-represented groups.
The Information and Computer Science department at UC Irvine will be
awarding over 20 graduate fellowships to US applicants for PhD
graduate study in the coming year. The fellowships include full
tuition and stipend.
The department is very active in ML and KDD-related research. Its
faculty include Dennis Kibler and Michael Pazzani (machine learning and
data mining), Padhraic Smyth (probabilistic learning and graphical
models), Rina Dechter (probabilistic and constraint-based reasoning),
Rick Lathrop (computational biology), Rick Granger (computational
neuroscience), Sharad Mehrotra (databases and multimedia information
retrieval) and Wanda Pratt (medical informatics and information
access).
The deadline for graduate student applications for Fall 1999
is January 15th. Full details are available online at http://www.ics.uci.edu/~chair/phd.html
MS DEGREE IN KNOWLEDGE DISCOVERY IN DATA
The goal of this MS degree program is to educate students in both the
fundamental principles of computational methods for modeling data as
well as a providing a broad foundation in emerging methods for
knowledge discovery and data mining. Technological advances in digital
data collection, memory capacity, and computational power, have
revolutionized our view of data analysis in the past 10 years. The
volumes of data being collected in science, business, medicine, and
government are truly vast in nature. Across all of these areas, there
is a rapidly increasing demand for better theories and tools to
provide users with improved understanding of their data and to
leverage their data for decision support.
There is some support available for M.S. students as internships in
industrial laboratories. The program (summarized below) consists of
courses in CS, AI and statistics. Full details are available
online at http://www.ics.uci.edu/~chair/ms.html
Previous11NextTop
Date: Tue, 17 Nov 1998 11:21:26 -0500
From: Trish Carbone carbone@mitre.org
Subject: Federal Data Mining Symposium & Exposition '99
The Federal Data Mining Symposium & Exposition '99 Call for Papers brochure
is now available. Attached is a copy of that announcement. Start now on
planning your paper, exhibit, or overall participation in this Must-See
Event!!
Look forward to seeing you there,
Trish Carbone, The MITRE Corporation
Conference Chair
Michelle Japzon, AFCEA International
Conference Coordinator
**By the way, if you have any questions, Michelle can be reached at events@afcea.org
(Attn: Michelle) or 703-631-6128.
CALL FOR PAPERS
FEDERAL DATA MINING SYMPOSIUM & EXPOSITION '99
McLean Hilton/7920 Jones Branch Drive/McLean, VA
March 9-10, 1999
'Data Mining Technology and Applications in the Government Community'
AFCEA and participating federal agencies, proudly present the second annual
Federal Data Mining Symposium and Exposition '99. Based on the success of
last year's symposium, the theme for this year will be Data Mining
Technology and Applications in the Government Community, emphasizing the
need for better and more automated methods of analysis. As the amount of
data being collected and stored increases dramatically, it becomes
imperative that we prepare for the future, not just within the Government
Community, but also within any community wherein data mining preparation
will facilitate the creation of knowledge from which future decision-making
may take place.
Data users, analysts, administrators, managers, developers, researchers,
theoreticians, and vendors are cordially invited to attend and submit
papers for presentation at Federal Data Mining '99. The 1st Federal Data
Mining Symposium included numerous vendors and all types of researchers and
users of data mining tools and techniques to create a unique opportunity to
discuss data mining in the domain of the government. There is no element
of the Federal Government, nor information technology corporation that does
not have a critical interest in 'mining the golden nugget' from the vast
repositories of information available to them.
SCHEDULE FOR SUBMISSION:
November 30, 1998: Prospective presenters will submit two abstracts
(approximately 500 words) with a brief biographical sketch of the author(s)
to the AFCEA Programs Office and to MITRE, attn: Trish Carbone. We prefer
to receive the abstracts via e-mail at events@afcea.org
and carbone@mitre.org.
For details please contact organizers above
[edited GPS]
EURO-PAR'99
Toulouse, France
August 31 - September 3, 1999
Euro-Par is the premier European conference on parallel computing and
normally attracts about 300 participants. It is an annual
international conference, dedicated to the promotion and advancement
of all aspects of parallel computing. The objective of Euro-Par is
to provide a forum to promote the development of parallel computing
both as an industrial tool and as an academic discipline, extending
the frontiers of the state of practice as well as the state of the
art.
Euro-Par'99 features 23 topics. Each topic (formerly called workshop)
is arranged by a small committee, consisting of a global chair, a local
chair, and usually two, sometimes more vice-chairs. One main topic of
Euro-Par'99 is Parallel Data Mining and Knowledge Discovery.
Topic 22: HIGH-PERFORMANCE DATA MINING AND KNOWLEDGE
DISCOVERY
Programme Committee:
David Skillicorn, (Queen's University, Canada), Global Chair,
Vipin Kumar, (University of Minnesota, USA), Vice-Chair
Hannu Toivonen, (University of Helsinki, Finland), Vice-Chair
Domenico Talia, (ISI-CNR, Rende, Italy), Local Chair
Second Call for Papers: IAT'99
Hong Kong December 15-17, 1999
----------------------------
| Papers Due: May 31, 1999 |
----------------------------
SPONSORS
~~~~~~~
Hong Kong Baptist University
ACM Hong Kong
IEEE Hong Kong Section - Computer Chapter
INVITED SPEAKERS
~~~~~~~~~~~~~~~
Setsuo Ohsuga (Waseda University, Japan)
Jeffrey Bradshaw (The Boeing Company, USA)
Dan Ling (Microsoft Corporation, USA)
Jan Zytkow (University of North Carolina, USA)
The Asia-Pacific Conference on Intelligent Agent Technology (IAT) is
a high-quality, high-impact biannual agent conference series. As the
first meeting in this new series, IAT'99 will primarily focus on (i)
the state-of-the-art in the development of intelligent agents and (ii)
the theoretical and computational foundations of intelligent agent
technology. The aim of IAT'99 is to bring together researchers and
practitioners from diverse fields, such as computer science,
information technology, business, education, human factors, systems
engineering, and robotics to (i) examine the design principles and
performance characteristics of various approaches in intelligent agent
technology, and (ii) increase the cross-fertilization of ideas on the
development of autonomous agents and multiagent systems among
different domains. By encouraging idea-sharing and discussions on the
underlying logical, cognitive, physical, and biological foundations as
well as the enabling technologies of intelligent agents, IAT'99 is
expected to stimulate the future development of new models, new
methodologies, and new tools for building a variety of embodiments of
agent-based systems.