|October 11, 2005 Morning
Tutorial I:"Online Mining Data Streams: Problems, Applications and Progress"
Jian Pei (Simon Fraser University, Canada)
Haixun Wang (IBM T.J. Watson Research Center)
Philip S. Yu (IBM T.J. Watson Research Center)
Presenter: Jian Pei
|Abstract.In many emerging data intensive applications, including applications in sensor networks, stock market analysis, network communication management and intrusion detection, a tremendous volume of data arrives in the form of continuous streams. Online mining of data streams for knowledge discovery has become a novel and rapidly growing research direction in the last couple of years. Recently, a few exciting results have been published in this area, while at the same time, even more challenging problems have been identified. The tutorial will present a brief overview of the inherent challenges in mining data streams, a survey on the latest results in this line of research, and an introduction to some real-life applications.
Bio of presenter:
Jian Pei received the Ph.D. degree in Computing Science from Simon Fraser
University, Canada, in 2002, under Dr. Jiawei Han's supervision. He also
received the B. Eng. and the M. Eng. degrees, both in Computer Science,
from Shanghai Jiao Tong University, China, in 1991 and 1993, respectively.
He is currently an Assistant Professor of Computing Science at Simon
Fraser University, Canada. In 2002 - 2004, he was an Assistant Professor
of Computer Science and Engineering at the State University of New York
(SUNY) at Buffalo, USA.
His research interests can be summarized as developing effective and
efficient data analysis techniques for novel data intensive applications.
Particularly, he is currently interested in various techniques of data
mining, data warehousing, online analytical processing, and database
systems, as well as their applications in bioinformatics. His current
research is supported in part by the Natural Sciences and Engineering
Research Council of Canada (NSERC) and the National Science Foundation
(NSF) of the United States.
Since 2000, he has published over 60 research papers in refereed journals,
conferences, and workshops, has served in the organization committee and
the program committee of over 50 international conferences and workshops,
and has been a reviewer for leading academic journals. He is a member of
the ACM, the ACM SIGMOD, the ACM SIGKDD, and the IEEE Computer Society.
| October 11, 2005 Afternoon 14:40-17:40
Tutorial II:"Reasoning about Web Information Systems"
Klaus-Dieter Schewe1 (Massey University, New Zealand)
Bernhard Thalheim (Christian Albrechts University Kiel,Germany)
Presenter: Klaus-Dieter Schewe1
The tutorial addresses reasoning problems in Web Information Systems (WISs) that arise from the desire to personalise such systems to the various needs of their users, to adapt systems to different environments and access channels, to ensure that the service expected from the systems is delivered, and to guarantee consistency and integrity of the systems.
For this the tutorial first presents of overview of the co-design approach to WIS design, which specifies systems on different levels of abstraction. The tutorial will focus on two of these levels emphasising views of WISs as dialogue systems and as data-intensive applications, respectively. The first of these views leads to modelling a story space describing how users
navigate through the system, actors, i.e. clases of users, and tasks. The story space gives rise to an assignment-free process algebra, while actors define obligations and rights and rights of roles in a propositional deontic logic as well as preference rules on the story space that indicate different user profiles. On a lower level of abstraction the co-design approach emphasises content and functionality modelling, which requires support by views on some underlying database and operations on these views. Furthermore, the tutorial will discuss two more extensions dealing with hierarchies that enable different presentation granularity,
and cohesion that permits adaptivity to technical restrictions such as channel bandwidth or end-devices.
The story algebras can in fact be represented by Kleene algebras with tests (KATs), which enable a simple form of system personalisation. Preferences and goals of users that are modelled by equations on the KAT can be exploited for simple, but effective term rewriting for the purpose of simplifying the story space according to a user's needs. The propositional deontic logic can be used to reason about tasks, for which proof obligations can be formulated. Using higher-order dynamic logic the operations on the views give rise to various proof obligations for consistency with respect to static and dynamic constraints,personalisation
and satisfiability of goals.
Bio of presenter:
Klaus-Dieter Schewe (MSc, PhD, DSc) is full professor at Massey University in New Zealand and Director of Massey¡¯s Information Science Research Centre. His major research interests are database theory and systems, logic in databases and systems development methodologies, in particular for web information systems. He has published more than 140 refereed publications, and has been programme committee chair for several international events such as ADC, FoIKS and QSIC.