NCE/IRIS
(Networks of Centres of Excellence/Institute for Robotics and Intelligent
Systems) Phase IV
Continuing Project from Phase III
last updated: April 12th, 2002
Project Leader:
Dr.
Laks V. S. Lakshmanan, Department of
Computing Science, University of British
Columbia
Duration: 3 years (April 2002 - March 2005)
Level of Support:
Summary: The requested funding level is $175,000/year for three years (April
1, 2002 to March 31, 2005) of which $25,000 cash support will come from IBM
Canada and $150,000 from NCE. The level of support requested from NCE
remains the same as with the previously concluded IRIS-3 project.
Principal Investigators:
- Dr.
Laks V. S. Lakshmanan, Department of
Computing Science, University of
British Columbia
- Dr. Alberto O.
Mendelzon, Department of Computer
Science, University of Toronto
- Dr.
Raymond T. Ng, Department of Computer
Science, University of British
Columbia
- Dr. Ke Wang,
School of Computing Science,
Simon Fraser University
- Dr. Jiawei Han,
University of Illinois. On leave
from: Computing Science,
Simon Fraser University
Industry Researcher and Industry Member
Statement of Problem:
| Mission Statement: Opening
new frontiers in data mining, by developing foundations and technologies for
querying, analyzing, and mining novel forms of data such as data on the web,
for mining novel kinds of patterns -- eg., analyzing XML data to discover
schema and analyzing transactional data to discover actionable rules which
have a more direct impact on business -- and for unifying the apparently
different fields of data mining and OLAP. |
Objectives of the Project:
Based on our goal, the project is divided into the following five
subprojects.
1. Construction of Distributed Data Warehouses on
the Internet.
2. Techniques for Mining Actionable Business
Rules.
3. Integration of OLAP and Data Mining
Technologies.
4. Algorithms for Classifying XML Documents.
5. High Performance Data Mining.
For each subproject, we plan to investigate four major aspects:
fundamental principles, implementation methods, performance improvements,
and industry applications.
Project Milestones:
The three-year milestones, of all subprojects put together, are outlined as
follows.
Year 1 (April 2002 - March 2003).
- Internet-based distributed data warehouse construction: (1) study the
foundations of query languages and query optimization techniques appropriate
for distributed warehouses; (2) identify, with the aid of our industrial
partner, a distributed data warehousing application for later empirical
evaluation; and (3) evaluate how existing tools can facilitate the building of
distributed warehouses.
- Implement existing techniques for computing iceberg queries and for
constraint-based queries. Evaluate how to adapt various outlier
detection techniques to OLAP cubes.
- Develop algorithms for classifying XML documents.
Year 2 (April 2003 - March 2004).
- Develop and evaluate alternative architectures for realizing distributed
data warehouses.
- Develop parallel algorithms for constraint-based queries. Develop
outlier detection algorithms for OLAP cubes.
- Develop FP-tree based algorithm and SSM-based algorithm for dynamic
constraint-based queries.
- Develop algorithms for mining schema for given XML document collections
and for mining actionable rules from data sets.
Year 3 (April 2004 - March 2005).
- Implement a prototype distributed data warehouse for a real application
and benchmark its performance.
- Evaluate and optimize parallel algorithms for constraint-based queries.
Implement a prototype module for detecting outlying cells in OLAP cubes.
- Investigate distributed variants of mining algorithms developed and
integrate with the distributed warehouse built above.
Project Team:
- Han, Jiawei - University of Illinois (currently on leave
from Simon Fraser University)
- Knorr, Ed - University of British Columbia
(PhD)
- Lakshmanan, Laks - University of British Columbia
- Ng, Raymond - University of British Columbia
- Zhou, Xiaodong - University of British Columbia (MSc)
-
Afshar, Ramin - Simon Fraser University (MSc)
-
Belchev, Eugene - Simon Fraser University (MSc)
-
Jin, Win - Simon Fraser University (PhD)
-
Pei, Jian - Simon Fraser University (PhD)
- Wang, Ke - Simon Fraser University
- Zheng, Yvonne - Simon Fraser University (MSc)
-
Barbosa, Denilson - University of Toronto (PhD)
- Mendelzon, Alberto - University of Toronto
-
Mignet, Laurent - University of Toronto (postdoc)
-
Pilar, Cecilia - University of Toronto (MSc)
-
Pu, Ken - University of Toronto (PhD)
-
Rizzolo, Flavio - University of Toronto (PhD)
-
Rogers, Yidan - University of Toronto (MSc)
-
Truta, Ramona - University of Toronto (MSc)
Publications:
UBC
Publications
U of T Publications
SFU Publications
M.Sc. Students Graduated:
from Concordia -
from SFU -
- Shi (Stone) Cong
- Haiming Huang
- Julia Itskevitch
- Joyce Man Lam
- Nancy Yaqin Liao
- Yiwen Yin
- Runying Mao
- Zhaoxia Wang
- Helen Pinto
- George Wenmin Li
- Behzad Mortazavi-Asl
- Benjamin Xuebin Lu
- Sonny H. S. Chee
software engineer, Pivotal Technology
Inc, Vancouver B.C., Canada -
- Jean Fen-ju Hou
- Jin Li
- Wei Wang
- Hua Zhu
- Yin Jenny (Chiang) Tam
- Gabor Melli
- Shan Cheng
from University of Toronto -
- G.O. Arocena
- B. He
- T. Palpanas
- A. Barta
- M. Wong
- F. Rizzolo
from University of British Columbia -
Ph.D. Students Graduated:
from University of Toronto -
- J.M. Turull
- D. Rafiei
- G.A. Mihaila
from SFU -
- Anthony K. H. Tung
- Krzysztof Koperski
- Osmar R. Zaine
- Yongjian Fu