CFP

Previous | item21 | Next

                    C A L L   F O R  P A P E R S

       Post-Processing in Machine Learning and Data Mining:
  Interpretation, Visualization, Integration, and Related Topics


                    :::::A WORKSHOP WITHIN::::

                             KDD-2000:
         The Sixth ACM SIGKDD International Conference on
                Knowledge Discovery and Data Mining

                August 20-23, 2000, Boston, MA, USA

http://www.acm.org/sigkdd/kdd2000



This workshop addresses an important aspect related to the Data Mining (DM) and
Machine Learning (ML) in post-processing and analyzing knowledge bases induced
from real world databases.

Results of a genuine ML algorithm, such as a decision tree or a set of decision
rules, need not be perfect from the view of custom or commercial applications.
It is quite known that a concept description (knowledge base, model) discovered
by an inductive process has to be usually processed by a post-pruning procedure.
Most existing procedures evaluate the extracted knowledge, visualize it, or
merely document it for the end user. Also, they may interpret the knowledge and
incorporate it into an existing system, and check it for potential conflicts
with previously derived knowledge (models). All post-processing procedures thus
provide a kind of "symbolic filter" for noisy, imprecise, or "non-user-friendly"
knowledge derived by an inductive algorithm.

The workshop will focus on all aspects of post-processing procedures, including:

o knowledge evaluation
o rule quality processing and evaluation
o knowledge filtering: rule truncation and postpruning
o interpretation of the knowledge (models) acquired
o explanation
o visualization of the new knowledge
o merging new knowledge with a-priori knowledge provided by a human expert
o knowledge combination and integration
o model refinement
o model fusion
o selecting the best presentation approach


Thus, the post-processing tools are complementary to the DM algorithms and
always help the DM algorithms to refine the acquired knowledge.  Usually, these
tools exploit techniques that are not genuinely logical, e.g., statistics,
neural nets, and others.

These reasons let us to launch this workshop. In fact, we would like to invite
practical, experienced, and empirical applications. We would like to focus on
industry and business applications, but we will review any functional
applications of the above concern in any discipline.

The theme of this workshop is directly related to applications of existing tools
in KDD:

1. The workshop will provide a forum for researchers and practitioners who are
   interested in the scope of the workshop to exchange information by attending
   and/or presenting a paper.  It is intended to encourage informal
   presentations of important problems, work in progress, or research ideas that
   may not have reached maturity.
2. The workshop organizers are interested in presentations from real world
   applications, vendors of data mining tools, specially focussing on
   challenges, limitations, constraints that they have faced in developing,
   using post-processing techniques and model fusion.
3. We have already organized within ICML-99 a workshop on pre- and
   post-processing.  Since majority of contributions at that workshop were
   focused on pre-processing only, we would like to support the post-processing
   issue within this KDD workshop.



Organizers
----------

A. (Fazel) Famili      (co-chair)             Ivan Bruha          (co-chair)
Editor-in-Chief, Intelligent Data Analysis    Dept. Computing & Software
Institute for Information Technology          McMaster University
National Research Council of Canada           Hamilton, Ont
Ottawa, Ont                                   Canada  L8S 4L7
Canada  K1A 0R6
email: Fazel.Famili@iit.nrc.ca email: bruha@mcmaster.ca
http://www.iit.nrc.ca/~fazel http://www.cas.mcmaster.ca/~bruha
phone: +1-613-9938554                         phone: +1-905-5259140 ext 23439
fax  : +1-613-9527151                         fax:   +1-905-5240340


Programming Committee
---------------------

Petr Berka, Laboratory of Intelligent Systems, University of Economics, Prague,
   Czech Republic
email: berka@vse.cz , http://lisp.vse.cz/~berka
Marko Bohanec, Institute Jozef Stephan, Jamova 37, Ljubljana, Slovenia
email: marko.bohanec@ijs.si , http://www-ai.ijs.si/MarkoBohanec/mare.html
Ivan Bruha   (co-chair)
A. (Fazel) Famili  (co-chair)
W.F.S. (Skip) Poehlman, McMaster University, Hamilton, Canada
email: skip@church.cas.mcmaster.ca



Organization Notes
------------------

There will be one or two invited talks on the workshop which will survey the
given topic as well as introduce their own research.

Each paper will be reviewed by two reviewers and up to 10 accepted papers will
be presented (each 15-20 min). If there is a larger interest, then some papers
might be accepted as posters.  Maximum size is 10 (ten) pages. Submit your paper
either by regular mail or by Email to both of the co-chairs.
                                   ^^^^^^^^^^^^^^^^^^^^^^^^
If you use Email, then please submit your paper in PostScript, format.

Workshop proceedings will be published as a technical report in collaboration
with the KDD-2000 organization committee.

Please note that authors of the best papers will be invited to submit an
extended version of their papers to the Intelligent Data Analysis Journal:
see http://www.iospress.nl/html/1088467X.html for details.


Schedule for paper submission
-----------------------------

Deadline for paper submission:          15 May 2000
Notification of acceptance:             15 June 2000
Deadline for final camera ready papers: 15 July 2000

Previous | item21 | Next