|
|
|
|
|
| Latest News
|
[April
11th, 2009] Confirmed Invited Speakers: Prof.
Lenore Mullin (NSF Program
Director and SUNY Albany) and Prof.
James Raynolds (SUNY Albany). Talk
title: Tensors and n-d Arrays: Mathematics of Arrays, Psi-Calculus, and Composition
of Tensor and Array Operations.
[April 13th, 2009]
Confirmed Invited Speaker: Prof. Charles Elkan (University
of California, San Diego). Talk title:
Factorizing
Matrices with Missing Entries: Alternative Approaches.
[April 15th, 2009]
Confirmed Invited Speaker: Prof.
Leiven De Lathauwer (Katholieke
Universiteit Leuven, Belgium); Dr. **Lathauwer** is the
developer of HOSVD. Talk title: Tensor
Decompositions and Applications: a Survey.
[April
20, 2009] Due to many requests, the deadline for paper submission has been
extended to May 5th, 2009.
[May15th, 2009]
The notification data has been
extended to May 25th, 2009.
[June
1st, 2009] Workshop Proceedings is posted
online.
 [June
16th, 2009] Workshop Program is posted
online.
|
| Workshop Description
|
|
This workshop is a continuation of the theme of
SIGKDD 2008 Workshop on Data Mining using Matrices and Tensors (DMMT’08).
DMMT’08 is the first workshop on data mining using matrices and tensors held
annually with the SIGKDD Conference. Our 2008 workshop was indeed a success and
more than 100 people attend the workshop.
The
field of pattern recognition, data mining and machine learning increasingly
adapt methods and algorithms from advanced matrix computations, graph theory and
optimization. Prominent examples are spectral clustering, non-negative matrix
factorization, Principal component analysis (PCA) and Singular Value
Decomposition (SVD) related clustering and dimension reduction, tensor analysis, L-1 regularization, etc. Compared to
probabilistic and information theoretic approaches, matrix-based methods are
fast, easy to understand and implement; they are especially suitable for
parallel and distributed-memory computers to solve large scale challenging
problems such as searching and extracting patterns from the entire Web. Hence
the area of data mining using matrices and tensors is a popular and growing area
of research activities.
This workshop will
present recent advances in algorithms and methods using matrix and scientific
computing/applied mathematics for modeling and analyzing massive,
high-dimensional, and nonlinear-structured data. One main goal of the workshop
is to bring together leading researchers on many topic areas (e.g., computer
scientists, computational and applied mathematicians) to assess the
state-of-the-art, share ideas, and form collaborations. We also wish to attract
practitioners who seek novel ideas for applications. In summary, this workshop
will strive to emphasize the following aspects:
- Presenting
recent advances in algorithms and methods using matrix and scientific
computing/applied mathematics
- Addressing the
fundamental challenges in data mining using matrices and tensors
- Identifying
killer applications and key industry drivers (where theories and
applications meet)
- Fostering
interactions among researchers (from different backgrounds) sharing the same
interest to promote cross-fertilization of ideas.
- Exploring
benchmark data for better evaluation of the techniques
|
|
Topics Areas
Top
|
|
Topic
areas for the workshop include (but are not limited to) the following:
Methods and algorithms:
- Principal
Component Analysis and Singular value decomposition for clustering and
dimension reduction
- Nonnegative
matrix factorization for unsupervised and semi-supervised learning
- Spectral graph
clustering
- L-1
Regularization and Sparsification
- Sparse PCA and
SVD
- Randomized
algorithms for matrix computation
- Web search and
ranking algorithms
- Canonical
Decompositions (CANDECOMP/PARAFAC)
-
Tensor analysis: Rank-1
Decomposition, PARAFAC/CANDECOMP, GLRAM/2DSVD,
Tucker
decompositions (e.g., the Higher-Order SVD) -
GSVD for
classification
- Latent Semantic
Indexing and other developments for Information Retrieval
- Linear,
quadratic and semi-definite Programming
- Non-linear
manifold learning and dimension reduction
- Computational
statistics involving matrix computations
- Feature
selection and extraction
- Graph-based
learning (classification, semi-supervised learning and unsupervised
learning)
Application areas
Information search
and extraction from Web
Text processing
and information retrieval
Image processing
and analysis
Genomics and
Bioinformatics
Scientific
computing and computational sciences
Social Networks
|
| Important Dates
Top |
-
April 20, 2009 May 5th, 2009: Electronic submission of full papers
-
May 15th, 2009 May 25th, 2009: Author notification
-
May 20th, 2009 May 29th, 2009: Submission of Camera-ready papers
- June 28th, 2009: Workshop in
Paris, France
|
| Paper Submissions
Top |
The electronic submission web site for research papers is available at:
http://www.easychair.org/conferences/?conf=dmmt09.
Please register at Easychair first if you did not use EasyChair before. Papers should be at most 10 pages long, single-spaced, in KDD conference
format, in font size 10 or larger with 1-inch margins on all sides.
|
| Program Committee Members
(tentative) Top |
-
Tammy Kolda, Sandia National Labs
-
Jesse Barlow, Penn State University
-
Michael Berry, University of Tennessee
-
Yun Chi, NEC Laboratories America
-
Lars Elden, Linkping University, Sweden
-
Christos Faloutsos, Carnegie Mellon University
-
Estratis Gallopoulos, University of Patras
-
Joydeep Ghosh, University of Texas at Austin
-
Ming Gu, University of Califonia, Berkeley
-
Michael Jordan, University of California, Berkeley
-
Huan Liu, Arizona State University
-
Michael Ng, Hong Kong Baptist University
-
Haesun Park, Georgia Tech
-
Wei Peng, Xerox Research
-
Robert Plemmons, Wake Forest
-
Alex Pothen, Old Domino University
-
Yousef Saad, University of Minnesota
-
Horst Simon, Lawrence Berkeley National Laboratory
-
Gang Wang, Microsoft Research
-
Fei Wang, Florida International University
-
Kai Yu, NEC Laboratories America
-
Hongyuan Zha, Georgia Tech
-
Zhongyuan Zhang, Central University of Finance & Economics
-
Shenghuo Zhu, NEC Laboratories America
|
|