News: *
GPS, what is new in KDNuggets *
Dan Rasmussen, Question: Image Processing Software Publications: *
D. Fisher, Book: Construction and Assessment of Classification Rules *
P. Turney, Hypertext Bibliographies on Machine Learning
http://www.ovum.com/evaluate/dmi/dmi000 Siftware: *
Dagmar Gerigk, WINROSA: automatically generates fuzzy rules from data *
R. Paulsen, Data Mining solutions for Call Centers Positions: *
D. Eide, Statistician position in Southern Connecticut *
Russ Greiner, PostDoc - Learning, Bayesian Nets - UofAlberta Meetings: *
R. Rajkumar, WSC3: World Conference on Soft Computing,
21-30 June 1998, On the Web,
Submissions are most welcome and should be emailed, with a
DESCRIPTIVE subject line (and a URL) to gps.
Please keep CFP and meetings announcements short and provide
a URL for details.
To subscribe, see www.kdnuggets.com/news/subscribe.html
KD Nuggets frequency is 2-3 times a month.
Back issues of KD Nuggets, a catalog of data mining tools
('Siftware'), pointers to Data Mining Companies, Relevant Websites,
Meetings, and more is available at Knowledge Discovery Mine site
at http://www.kdnuggets.com/
-- Gregory Piatetsky-Shapiro (editor)
gps
********************* Official disclaimer ***************************
All opinions expressed herein are those of the contributors and not
necessarily of their respective employers (or of KD Nuggets)
*********************************************************************
~~~~~~~~~~~~ Quotable Quote ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
From The Vermonter's Guide to Computer Lingo, original author unknown
Disk Operating System: The equipment the Doc uses when you have a
floppy disk.
RAM: The hydraulic thingy that makes the woodsplitter work.
Hard Drive: Gettin' home in mud season.
Prompt: What you wish the mail was in mud season.
Windows: What to shut when it's 30 below.
Screen: What you need for black fly season.
Byte: What black flies do.
Chip: What to munch on.
Micro Chip: What's left in the bag when the chips are gone.
Infrared: Where the left-over's go when Fred's around.
Modem: What you did to the hay fields.
Previous1NextTop
Date: Fri, 28 Nov 1997 09:41:10 -0500 (EST)
From: GPS (gps)
Subject: What is New in KDNuggets in November
Link Analysis: rp,
new entry for Belief Network Constructor,
a belief network learning system, which includes
a wizard-like interface and a construction engine.
under
Classification: Decision-tree: com
a new entry for Decisionhouse,
a suite of visualization and decision-tree tools
for scalable data analysis and sophisticated customer modelling.
under Multi-task Tools: com, updated entry for DataEngine 2.1
a software tool for intelligent data analysis, combining
combines conventional data analysis methods with
fuzzy technologies, neural networks and statistical methods.
under Human Resources,
new entry for Vacancy Matching System,
environment for matching and data mining in large
vacancy/applicant databases (up to 100k vacancies).
under Retail,
new link to
Knowledge Discovery One (KD1),
builds complete, sophisticated, yet easy-to-use applications that
allow retailers to better understand and predict their customers' buying
habits.
Previous2NextTop
Date: Mon, 17 Nov 1997 16:15:14 +0100 (MET)
From: 'Dan B. Rasmussen' (dan@ruc.dk)
Subject: image processing
Hi,
I am working on a project where I have to identify spots on a digitized
image, like FOCAS do in the SKICAT system. I am searching for literature and
C/C++ libraries or source code for image processing that can detect
contiguous pixels in the image that are to be grouped as one object.
I would like to know where I can obtain such information.
Previous3NextTop
Date: Sun Nov 16 16:23:26 1997
From: dfisher@vuse.vanderbilt.edu
(Douglas H. Fisher) via AI-STATS
Subject: BOOK: CONSTRUCTION AND ASSESSMENT OF CLASSIFICATION RULES
CONSTRUCTION AND ASSESSMENT OF CLASSIFICATION RULES
David J. Hand, John Wiley and Sons, 1997
ISBN 0-471-96583-9
CONTENTS:
PART I: BASIC IDEAS
1. Introduction
PART II: CONSTRUCTING RULES
2. Fisher's LDA and other methods based on covariance matrices
3. Nonlinear methods
4. Recursive partitioning methods
5. Nonparametric smoothing methods
PART III: EVALUATING RULES
6. Aspects of evaluation
7. Misclassification rate
8. Evaluating two class rules
PART IV: PRACTICAL ISSUES
9. Some special problems
10. Some illustrative applications
11. Links and comparisons between methods
This book is intended to be a comprehensive introduction to methods of
supervised classification, approached from the perspective that different
problems require different solutions. It covers all the main methods,
including: classical statistical ones such as linear and quadratic discriminant
analysis, structured covariance matrices, principal components regression,
partial least squares, SIMCA, DASCO, regularisation methods, shortest least
squares, logistic regression, neural networks, generalised additive models,
projection pursuit regression, radial basis functions, multivariate adaptive
regression splines, classification trees and graphs, and nonparametric nearest
neighbour and kernel methods. A special feature is its detailed treatment of
how to measure the performance of a classification rule - error rate is seldom
adequate in real problems. Many different measures of performance are described
and their properties examined. Detailed descriptions are given of supervised
classification methods applied in automated chromosome identification, credit
scoring, speech recognition, and character recognition.
========================
Professor David J. Hand
Dept of Statistics
The Open University
Milton Keynes
MK7 6AA
UK
Comments, corrections, and contributions are welcome.
- Peter Turney
Previous5NextTop
Date: Fri, 21 Nov 1997 10:41:00 +0000
From: Elisabeth Kyral (ebk@ovum.com)
Subject: New independent, detailed analysis of dm tools
Ovum has recently published an evaluation study of twelve leading data
mining tools. 'Ovum evaluates: Data Mining' guides you through data mining
and helps you formulate a plan for successful implementation. It examines
how others have used the technology and evaluates the current products. Each
evaluation contains 25 pages of fact, description and judgement.
To get more information about this report, please visit us at
New product for automatic
generation of fuzzy rules!
In 1997 MIT - Management of Intelligent Technologies from Aachen in
Germany started distributing WINROSA worldwide. WINROSA automatically
generates Fuzzy Rules from data.
Compared to conventional techniques fuzzy methods have offered
superior solutions in numerous applications. Nevertheless, generating
the desired rule base has turned out as to be time-consuming and
difficult. Exactly here WINROSA will support you!
This software tool is based on the Fuzzy-ROSA-method, that has been
developed under supervision of Prof. Dr. Harro Kiendl at the
University of Dortmund, Germany. Prof. Kiendl comments on the new
generation of fuzzy software tools: 'WINROSA allows the automatic
generation of Fuzzy Rules that are based on data collected from
processes and observations'.
Karl Lieven, Managing Director of Management of Intelligent
Technologies emphasizes the economic advantages of using WINROSA by
mentioning: 'WINROSA generates Fuzzy Rules that can be read in and be
processed by leading fuzzy inference tools like e.g. DataEngine(R),
fuzzyTECH(R), MatLab(R) and Dora-Fuzzy. By that means productivity as
well as application performance are improved. This leads to a very
fast return on investment.'
WINROSA: The advanced tool for creating fuzzy systems and also for
their application!
For further information on WINROSA please contact:
Dr. Richard Weber, MIT - Management of Intelligent Technologies
Promenade 9, 52076 Aachen, Germany
Telephone: +49 / 24 08 / 9 45 80; Telefax: +49 / 24 08 / 9 45 82
E-mail: rw@mitgmbh.de;
Previous7NextTop
From: 'Paulsen, Robert A' (Robert.A.Paulsen@siemenscom.com)
Subject: Data Mining solutions for Call Centers
Date: Wed, 26 Nov 1997 12:41:41 -0800
Siemens Communications, the world leader in communications equipment,
and Sabre Technologies have partnered to create a comprehensive call
center data mining solution. Initially designed to meet the needs of the
Utilities and Communications markets as they prepare for deregulation,
the solution has applications across industries. By positioning customer
service reps with such in depth information, including predictions for
customer interests, companies can transform their call center into a
proactive marketing tool.
Bob Paulsen
Siemens Business Communication Systems - Energy & Communications
Services
6455 South Yosemite St., Suite 700
Englewood, CO 80111
(303) 773-7625
Previous8NextTop
From: dje@dmc22.com
Date: Wed, 19 Nov 1997 22:22:56 -0500 (EST)
Subject: Statistician Needed in Southern Connecticut
I am looking to hire a Statistician to participate in the design and analysis of clinical trials
and write statistical reports which summarize the methodology and results of the trials. After
review, these reports are then submitted to the FDA as a crucial part of our new drug
applications (NDA) for each product.
The incumbent must possess an advanced degree and a thorough/expert knowledge of a wide
range of Statistical methodology including experimental design, linear models, categorical
data techniques, non-parametric statistics and survival analysis, and must display a firm
understanding of advanced statistical probability theory. Knowledge of computer
programming, particularly statistical software packages (preferably SAS) is essential.
The position is in southern Connecticut and offers permanent employment with one of the
world's largest and most recognized companies. Excellent benefits and salary from $50,000
to $90,000, depending on the level of experience, make this an outstanding opportunity for
the right person.
If you know someone that would be interested I can be contacted at:
Dave Eide
Voice: (609) 584-9000 ext. 273
Fax: (609) 584-9575
POST-DOCTORAL RESEARCH FELLOWSHIP
IN COMPUTER SCIENCE
University of Alberta
Edmonton, Canada
Applications are invited for a one-year (renewable) fellowship to work
in the areas of
* machine learning / learnability / datamining
* knowledge representation, especially Bayesian networks and other
probabilistic structures.
Candidates should have a PhD in Computer Science or the equivalent,
and will be required to carry out high quality research, to obtain
both theoretical and empirical results. Previous research excellence
and strong productivity in addition to good computing background is
essential.
Applications including
* CV
* statement of interests
* 1 or 2 publications
* list of references
should be sent ASAP (but no later than 15 January 1998) to:
Russell Greiner
Department of Computing Science
615 General Service Bldg
University of Alberta
Edmonton, AB T6G 2H1
Previous10NextTop
Date: Tue, 18 Nov 97 21:22:27
From: 'Roy, Rajkumar' (RRoy@cim.cran.ac.uk)
Subject: WSC3: 2nd CFP
Second Call for Papers and On-line Tutorials
3rd On-line World Conference on Soft Computing
in Engineering Design and Manufacturing
(WSC3)
21-30 June 1998
On the Internet (World-Wide Web)
Hosted by:
Cranfield University, United Kingdom
University of Bath, United Kingdom
Nagoya University, Japan
Michigan State University, USA
University of Cape Town, South Africa
---------------------------------------------------------------------------
WSC3 Servers:
Original papers and on-line tutorial proposals are invited for the 3rd
Online World Conference on Soft Computing (WSC3) to be held on the
Internet. WSC3 will bring together practitioners and researchers in
soft computing in engineering design and manufacturing across the world
with the aim to publish quality research rapidly and with less cost.
CO-SPONSORS:
IEEE Industrial Electronics Society (IES)
British Telecommunications Ltd. (BT), UK
AIT Centre, Cranfield University, UK
TOPICS OF INTEREST:
The scope of this conference covers the following soft computing and
related techniques and their application to engineering design and
manufacturing:
Fuzzy Logic
Neural Networks
Evolutionary Computing
Other Stochastic Optimisation Techniques
Hybrid Methods
Intelligent Agents and Agent Theory
Causal Models
Data Mining
Probabilistic Reasoning
Case-based Reasoning
Chaos Theory
Interactive Computational Models
****************
Please visit the WSC3 Servers for further details.
Previous11NextTop
From: Peter Bartlett (Peter.Bartlett@keating.anu.edu.au)
Subject: COLT98 call for papers
Date: Tue, 25 Nov 1997 18:34:09 +1100 (EST)
Web:
CALL FOR PAPERS: COLT '98
Eleventh Annual Conference on Computational Learning Theory
University of Wisconsin-Madison
July 24-26, 1998
The Eleventh Annual Conference on Computational Learning Theory
(COLT '98) will be held at the University of Wisconsin-Madison from
Friday, July 24 through Sunday, July 26, 1998.
The conference will be co-located with the Fifteenth International
Conference on Machine Learning (ICML '98) and the Fourteenth
Conference on Uncertainty in Artificial Intelligence (UAI '98).
Registrants to any of COLT, ICML, or UAI will be allowed to attend,
without additional costs, the technical sessions of the other two
conferences. Joint invited speakers, poster session, and a panel
session are planned for the three conferences. The conferences will
be directly followed by the Fifteenth National Conference on
Artificial Intelligence (AAAI '98). The AAAI tutorial and workshop
program will be held the day after the co-located conferences
(Monday, July 27), and we anticipate that this program will include
workshops and tutorials in the machine learning area. On the same
day, UAI will offer a full day course on uncertain reasoning. There
will be six other AI-related conferences held in Madison around this
time.
We invite papers in all areas that relate directly to the analysis of
learning algorithms and the theory of machine learning. Some of the
issues and topics that have been addressed in the past include:
* design and analysis of learning algorithms;
* sample and computational complexity of learning specific model
classes;
* frameworks modeling the interaction between the learner, teacher
and the environment (such as learning with queries, learning control
policies and inductive inference);
* learning using complex models (such as neural networks and decision
trees);
* learning with minimal prior assumptions (such as mistake-bound
models, universal prediction, and agnostic learning).
We strongly encourage submissions from all disciplines engaged in
research on these and related questions. Examples of such fields
include computer science, statistics, information theory, pattern
recognition, statistical physics, inductive logic programming,
information retrieval and reinforcement learning. We also encourage
the submission of papers describing experimental results that are
supported by theoretical analysis.