The 5th ICDM Workshop on High Dimensional Data Mining (HDM’17)
In conjunction with the IEEE International Conference on Data Mining (IEEE ICDM 2017)

New Orleans, USA, November 18, 2017.

Schedule

Morning session: Similarities and geometry
Chair:
Dixon Vimalajeewa

 9:15 – 9:30: Welcome

 9:30 – 10:00: A Novel Method for Fast and Accurate Similarity Measure in Time Series Field - Miaomiao Zhang, Dechang PI 

10:00 – 10:15: Coffee

10:15 – 10:45: Differential Geometric Retrieval of Deep Features - Y Qian, E Vazquez, and B Sengupta

10:45 – 11:15: Robust Projective Dictionary Learning by Joint Label Embedding and Classification - Weiming Jiang, Zhao Zhang, Jie Qin, Mingbo Zhao, Fanzhang Li, Shuicheng Yan

11:15 – 11:45: Multiple Queries of Information Retrieval using Krylov Subspace Method - Youzuo Lin

11:45 – 1:00: Lunch

Afternoon session: Dimensionality reduction and density estimation.
Chair: Amir Karami

 1:00 – 1:30: Lazy stochastic principal component analysis - Michael Wojnowicz, Dinh Nguyen, Li Li, Xuan Zhao

1:30 – 2:00: Near-optimal Noisy Low-tubal-rank Tensor Completion via Singular Tube Thresholding - Andong Wang, Zhong Jin

2:00 – 2:30: Taming Wild High Dimensional Text Data with a Fuzzy Lash - Amir Karami

2:30 – 3:00: Evaluation of Non-linearity in MIR Spectroscopic Data for Compressed Learning - Dixon Vimalajeewa, Donagh Berry, Eric Robson, Chamil Kulatunga

3:00 – 3:15: Coffee

3:15 – 3:45: High-dimensional Density Estimation for Data Mining Tasks - Alexander Kuleshov , Alexander Bernstein, and Yury Yanovich



Description of Workshop

 

Over a decade ago, Stanford statistician David Donoho predicted that the 21st century will be the century of data. "We can say with complete confidence that in the coming century, high-dimensional data analysis will be a very significant activity, and completely new methods of high-dimensional data analysis will be developed; we just don't know what they are yet." -- D. Donoho, 2000.

 

Unprecedented technological advances lead to increasingly high dimensional data sets in all areas of science, engineering and businesses. These include genomics and proteomics, biomedical imaging, signal processing, astrophysics, finance, web and market basket analysis, among many others. The number of features in such data is often of the order of thousands or millions - that is much larger than the available sample size.

For a number of reasons, classical data analysis methods inadequate, questionable, or inefficient at best when faced with high dimensional data spaces:

 1. High dimensional geometry defeats our intuition rooted in low dimensional experiences, and this makes data presentation and visualisation particularly challenging.

 2. Phenomena that occur in high dimensional probability spaces, such as the concentration of measure, are counter-intuitive for the data mining practitioner. For instance, distance concentration is the phenomenon that the contrast between pair-wise distances may vanish as the dimensionality increases.

3. Bogus correlations and misleading estimates may result when trying to fit complex models for which the effective dimensionality is too large compared to the number of data points available.

 4. The accumulation of noise may confound our ability to find low dimensional intrinsic structure hidden in the high dimensional data.

 5. The computation cost of processing high dimensional data or carrying out optimisation over a high dimensional parameter spaces is often prohibiting.

 

Topics

 

This workshop aims to promote new advances and research directions to address the curses and uncover and exploit the blessings of high dimensionality in data mining. Topics of interest include all aspects of high dimensional data mining, including the following:

- Systematic studies of how the curse of dimensionality affects data mining methods

- Models of low intrinsic dimension: sparse representation, manifold models, latent structure models, large margin, other?

- How to exploit intrinsic dimension in optimisation tasks for data mining?

- New data mining techniques that scale with the intrinsic dimension, or exploit some properties of high dimensional data spaces

- Dimensionality reduction

- Methods of random projections, compressed sensing, and random matrix theory applied to high dimensional data mining and high dimensional optimisation

- Theoretical underpinning of mining data whose dimensionality is larger than the sample size

- Classification, regression, clustering, visualisation of high dimensional complex data sets

- Functional data mining

- Data presentation and visualisation methods for very high dimensional data sets

- Data mining applications to real problems in science, engineering or businesses where the data is high dimensional

 

Paper submission

High quality original submissions are solicited for oral and poster presentation at the workshop. The page limit of workshop papers is 8 pages in the standard IEEE 2-column format (http://www.ieee.org/conferences_events/conferences/publishing/templates.html), including the bibliography and any possible appendices.

All papers must be formatted according to the IEEE Computer Society proceedings manuscript style, following IEEE ICDM 2017 submission guidelines available at http://icdm2017.bigke.org/. All submissions will be peer-reviewed, and all accepted papers will be included in the IEEE ICDM 2016 Workshops Proceedings volume published by IEEE Computer Society Press, and will also be included in the IEEE Xplore Digital Library. The workshop proceedings will be in a CD separated from the CD of the main conference. The CD is produced by IEEE Conference Publishing Services (CPS).


Important dates

·         Submission deadline: 14th August.

·         Workshop paper notifications: September 4, 2017

·         Camera-ready deadline for the final version of accepted papers: September 15, 2017

·         Workshop date: November 18th 2017

 

Registration & Expenses

Every workshop paper must have at least one full paid conference registration in order to be published. Check the main conference pages for details.

 

Program committee

Stephan Gunnemann - Carnegie Mellon University, USA 

Michael E. Houle - National Institute of Informatics, Japan 

Ata Kaban - University of Birmingham, United Kingdom 

Mehmed Kantardzic - University of Louisville, USA 

Mark Last - BGU, Israel 

Milos Radovanovic - University of Novi Sad, Serbia 

Nenad Tomasev - Google, Mountain View, CA, USA

Guoxian Yu - Southwest University, China 

 

Workshop organisation & contact

Dr. Ata Kaban

School of Computer Science, University of Birmingham, UK

 

Previous editions:

HDM’16

HDM’15

HDM’14

HDM’13

 

 

Related Links & resources: Analytics, Big Data, Data Mining, & Data Science Resources