Corpora: 2nd cfp KDAD'99

Mohamed Quafafou (Mohamed.Quafafou@irin.univ-nantes.fr)
Mon, 8 Mar 1999 17:13:49 +0100

Dear Colleagues,

..Profuse apologies for cross-posting. Please forward to anyone who would
be interested. Thank you ..

The PAKDD Workshop on Knowledge Discovery from Advanced Databases (KDAD'99)

In conjunction with

Third Pacific-Asia Conference on Knowledge Discovery and Data Mining
(PAKDD'99) Beijing, China, April 26, 1999

http://www.sciences.univ-nantes.fr/kdad99

MOTIVATION

The main goal of knowledge discovery is to convert the massive amount of
the captured data to actionable knowledge. The last few years have seen a
growing development of algorithms and approaches to automatic discovery of
knowledge from for well-structured raw data stored in very large relational
databases. However, different kind of data, i.e., image, text, video, etc.
are know available in file systems, objects servers and web sites and
accessed by a wide range of users, ranging from discipline experts to
novice users. There is a growing need for new generation of data mining
and KDD tools to automatically and intelligently analyzes complex data
types as audio, image, video, spatial data, temporal data and textual
information. The challenge for the next decade is to develop theoretical
foundations, technological support and integrated tools to make easy the
discovery of knowledge from advanced databases (KdAd) by converting its
heterogeneous information to useful and actionable knowledge according to a
users goals.

The problem of efficiently discovering knowledge from advanced databases is
an important problem in wide variety of application areas including
astronomy, geology, biology, urbanism, environment, biomedical, spatial
relationship marketing, etc. Knowledge discovery from advanced databases
development faces challenges in several issues, including data storage and
access, data preprocessing, data mining algorithms and architectures,
visualization and interactive exploration, interfaces and languages for
data mining, etc.:

>> 1. Complex data which may be extremely varied in content, structure and
>>format, can not be stored and manipulated in a traditional RDBMS.
>>Alternative ways are extending object-relational databases to store and
>>manage both data and methods encapsulated into objects, and objects
>>databases which allow a persistent storage of objects and an the use
object-oriented programming language.
>> 2. Large preprocessing efforts are generally necessary to understand the
>>data and data preprocessing includes different kind of operators as
>>indexing schemes, retrieval methods, data integration, metadata creation,
>>etc.
>> 3. Efficient data mining algorithms are needed to analyse this
>>heterogeneous data which may be strongly structured (i.e., multimedia
>>object) or poorly structured (i.e., informal notes added to an image).
>>The integration of KDD tools into software environment is primary of
>>importance in real world applications, i.e., coupling KDD tools with a
geographical information system. The KdAd99 workshop will be focused on
algorithms, methodologies, technologies and standards related to knowledge
discovery from advanced databases, i.e., OODB, text mining, spatial and
temporal data exploration and analysis, image, audio, video and more
generally multimedia objects mining. Real world applications are strongly
encouraged.

WORKSHOP OUTLINE

The Workshop includes invited and contributed talks which give a
description of open questions in KdAd, work in progress, solutions
presented by specific approaches, lessons learned from realized real world
projects. This workshop is a forum for discussion of new ideas and
techniques and will leads to identify the main problems which should be
addressed in the immediate future. We invite participants from academia,
government and industry to share ideas and experiences.

SUBMISSION OF PAPERS

Currently, researchers in different communities (i.e., data mining,
database, multimedia, digital libraries, metadata, data engineering and
object) are looking at different aspects of this problem. The primary
purpose of this workshop is threefold :

>> 1. to bring together researchers and practitioners interested by KdAd
>>problem in order to exchange ideas,
>> 2. to gain a better understanding of the state of the art and the
>>technological solutions,
>> 3. to identify and explicit the search challenges to address in getting
>>specific tools and/or integrated solutions supporting the emergence of
>>knowledge from Advanced databases.

Among others, papers of the following kind are welcome :
- Heterogeneous Database systems
- Semi-structured Data mining
- Temporal and Spatial Database systems
- Multimedia and Digital Libraries
- Image, Video and Audio Data exploration and analysis
- OODB mining
- Text Mining
- Metadata and complex object mining
- Agent-based architectures
- Analysis and methodology issues
- Applications in Sciences, Engineering, GIS-based mining, Internet
and web mining, etc.

This workshop addresses practitioners as well as researchers from those
communities which contribute to the KdAd topic. Potential attendees submit
a full technical paper (not exceeding 5000 words), or a summary of an
ongoing research effort (not exceeding 1500 words). Electronic submission
(postscript, pdf, or MS Word format) is highly encouraged.
Submission should be sent to quafafou@irin.univ-nantes.fr. Hard-copy
submission are also accepted, please sent three (3) copies of the paper to
Mohamed Quafafou (see address below)

IMPORTANT DATES

Submission deadline March 15, 1999
Notification of acceptance April 5, 1999
Workshop date April 26, 1999

REVIEW & PUBLICATION

All submissions will be reviewed on the basis of relevance, originality,
significance and clarity. Two referees will review each submission and
their results will be sent to the first author via email, unless requested
otherwise. Authors of selected papers will be invited to extend their
papers for inclusion in a special issue of An International Journal.

KDAD'99 ORGANISATION

Mohamed Quafafou (co-chair)
University of Nantes, 2, rue de la Houssiniere, BP. 92208 44322 Nantes
Cedex 3 France.
e-mail : quafafou@irin.univ-nantes.fr

Philip Yu (co-chair)
IBM T.J. Watson Research Center
30 Saw Mill River Road Hawthorne, NY 10532, USA
e-mail: psyu@watson.ibm.com

PROGRAM COMMITTE

Chung Sheng Li IBM T.J. Watson Research Center, USA
Hannu Toivonen University of Helsinki, Finland,
Raymond Ng Univ. of British Columbia, Canada
Kyusoek Shim Bell Labs, USA
Jiawei Han Simon Fraser University, Canada
Rakesh Agrawal IBM Almaden Research Center, USA
Hongjun Lu National University of Sangapore, Singapore
Vijay V. Raghavan University of Southwestern Louisiana, USA
Ahmed K. Elmagarmid Perdue University, USA
Ning Zhong Yamagushi university, Japan
Xindong Wu Colorado School of Mines, USA

-- Mohamed Quafafou

IRIN, 2 rue la Houssiniere, BP 92208 - 44322, Nantes cedex 03, France.
* tel: +33 (0) 251 125 853 * fax: +33 (0) 251 125 812 *
quafafou@irin.univ-nantes.fr *