Call for papers
There is a growing interest in learning data topology, from theoretical results to real-world applications. Topological learning is an emerging field which is expected to bring new insights in all the areas of the data mining and knowledge discovery process.
Data mining and knowledge discovery (KDD), as stated in their early definition, can today be considered as stable fields with numerous efficient methods and studies that have been proposed to extract knowledge from data. In every data mining task, understanding the structure of multidimensional patterns, in supervised or unsupervised cases, is of fundamental importance. Topological learning aims at finding hidden structures (usually low-dimensional manifolds) in order to better understand and exploit data.
The aim of this workshop is to address issues related to the concepts of learning data topology. Our goal is to attract papers dealing with each step of this field. Actually, learning data topology within the KDD process implies to work on every step, starting from the pre-processing (e.g. structuring and organizing) to the visualization and interpretation of the results, via the data mining methods themselves.
There are still big practical and theoretical challenges dealing with data topology, among which: how to deal with noisy, high-dimensional, multi-scale or complex data (times series, images, graphs, trees, texts, etc.)? How to interpret and qualify the results? How to select the models complexity? How to design efficient algorithms with theoretical guarantees? When is topological learning beneficial against usual methods? Etc. Papers dealing with these issues and reporting applications on real data will be of fundamental interest for this emerging field, and will take high priority for the selection process. KDD fields which involve topological learning, include, but are not limited to:
- Topology learning
- Manifold learning
- Spectral clustering and embedding
- Spectral feature selection
- Linear and non linear dimensionality reduction
- Computational topology
- Mathematical morphology
- Geometric Inference
- Topological learning and complex data
- Applications and experience feedback
More and more people approach the field of topological learning from different and interesting angles. They come from various communities such as data mining, statistics, geometry, topology, physics, medicine or engineering. We believe that now is the right time to establish and enhance cross fertilization between these communities. During this workshop, researchers will have the opportunity to bring new ideas, discuss their experiences and contribute to the theoretical and practical maturation of topological learning.
The workshop will consist in a series of communications (oral presentations or poster). A reasonable time will be left for the discussion after each presentation. All the articles will be reviewed at least twice with a goal to improve their quality and give advice to the authors. A dedicated place will be given to the young researchers with a session (Position paper) grouping the work in progress in the various European teams. That can be the occasion for a PhD student or a young researcher to present his/her starting project. This session will be particularly significant for work on the beginning and the installation of research groups on shared topics. Demonstrations of research results could be associated with the poster presentations.
Instructions for authors
The submissions should be written in English and should not exceed 10 pages in
the Springer Verlag format.
Submitted papers will be evaluated by at least three reviewers.
Any submission that exceeds length limits or deviates from formatting requirements
may be rejected without review.
||June 8, 2009 (Extended)
||June 15, 2009 (Extended)
( Passed )
||July 20, 2009
( Passed )
||July 31, 2009
( Passed )
||September 14, 2009
( Passed )