BISC Seminar Announcement, Monday August 17th, 4-5pm, 310 Soda

Frank Hoffmann (fhoffman@cs.berkeley.edu)
Sun, 16 Aug 1998 05:10:54 +0200 (MET DST)

************************************************************
B I S C S e m i n a r A n n o u n c e m e n t
************************************************************

Learning Possibilistic Networks: Data Mining Applications

Speaker : Rudolf Kruse
Department of Computer Science
University of Magdeburg
Email : Rudolf.Kruse@cs.uni-magdeburg.de

!!!!!!!! Notice this seminar takes place on a MONDAY !!!!!!!!!!
!!!!!!!! rather than the usual thursday !!!!!!!!!!

Date: Monday, August 17th, 1998
Time: 4-5pm
Location : 310 Soda Hall

Abstract

Since reasoning in multi-dimensional domains tends to be infeasible in
the domains as a whole --- and the more so, if uncertainty and/or
imprecision are involved --- decomposition techniques, that reduce the
reasoning process to computations in lower dimensional subspaces, have
become very popular. For example, decomposition based on dependence
and independence relations between variables has been studied
extensively in the field of graphical modeling. The best known
approaches are based on probability theory (e. g. Bayesian networks),
but recently possibilistic networks gained a lot of interest due to
their close connection to fuzzy methods.

In this talk the problem of learning possibilistic networks from
data is studied, i. e. how to determine from a database of sample cases an
appropriate decomposition of the possibility distribution on the domain
under consideration. Such automatic learning is important, since
constructing a network by hand can be tedious and time-consuming. If a
database of sample cases is available, as it often is, learning algorithms
can take over at least part of the construction task.

These new methods can be used to do "data mining" i. e. to discover
useful knowledge that is hidden in the large amounts of data stored in
data warehouses. Several industrial applications are studied, e. g. an
application for the automotive industry, in which the induction of
possibilistic networks was used to find weaknesses in Mercedes Benz
vehicles and thus to improve the product quality.

**********************************************************************
Please direct questions with regard to the contents of the talk
and request for papers to the speaker. To unsubscribe send email
to fhoffman@cs.berkeley.edu NOT !! to bisc-group@diva.eecs.berkeley.edu
**********************************************************************

-- 
---------------------------------------------------------------------------
Frank Hoffmann                               UC Berkeley
Computer Science Division                    Department of EECS
Email: fhoffman@cs.berkeley.edu              phone: 1-510-642-8282
URL: http://http.cs.berkeley.edu/~fhoffman   fax:  1-510-642-5775
---------------------------------------------------------------------------

############################################################################ This message was posted through the fuzzy mailing list. (1) To subscribe to this mailing list, send a message body of "SUB FUZZY-MAIL myFirstName mySurname" to listproc@dbai.tuwien.ac.at (2) To unsubscribe from this mailing list, send a message body of "UNSUB FUZZY-MAIL" or "UNSUB FUZZY-MAIL yoursubscription@email.address.com" to listproc@dbai.tuwien.ac.at (3) To reach the human who maintains the list, send mail to fuzzy-owner@dbai.tuwien.ac.at (4) WWW access and other information on Fuzzy Sets and Logic see http://www.dbai.tuwien.ac.at/ftp/mlowner/fuzzy-mail.info (5) WWW archive: http://www.dbai.tuwien.ac.at/marchives/fuzzy-mail/index.html