Everyday, huge amounts of data are generated by users via the web, social networks, etc.
Clustering/Coclustering techniques are a tool of choice to help organize the huge collections of data that increasingly beset us. Clustering/Coclustering
is an unsupervised learning approach that allows one to discover global structures in the data (i.e. clusters). Given a dataset, it identifies different data
subsets which are hopefully meaningful. The discovered clusters are deemed interesting if they are while instances within each (co)cluster share similar features.
This (co)clustering problem has motivated a huge body of work and has resulted in a large number of algorithms. (Co)Clustering has thus been used in numerous
real-life application domains such as marketing, city planning, and so forth.
Clustering algorithms are a tool of choice to explore these high-dimensional data sets. However numerous questions remain open as:
- What are the last advances in
- supervised clustering that combines the main characteristics of both traditional clustering and supervised classification tasks?
- quality criteria?
- clustering for big data?
- evolving clustering?
- clustering events or time series?
This special session offers a meeting opportunity for academics and industry researchers belonging to the communities of Computational Intelligence, Machine Learning, Experimental Design, and Data Mining to discuss new areas of (co)clustering, One goal of this special session will be two-fold: On the one hand, to look for new algorithms and techniques proposals based on (co)clustering. On the other hand, to look for new application domains, real problems, where the application of (co)clustering have demonstrated an outstanding performance or interpretation abilities against other traditional approaches.
Publication opportunities: Papers should be submitted to IJCNN. We encourage papers that describe new algorithm and applications of (co)clustering in real-world. In the industrial context, the main difficulties met and the original solutions developed, have to be described.
Paper acceptance and publication will be judged on the basis of their quality and relevance to the special session themes, clarity of presentation, originality and accuracy of results and proposed solutions.
The set of proposed topics includes, but is not limited to::
- Clustering, Coclustering
- Supervised Clustering (Coclustering)
- Semi-Supervised Clustering (Coclustering)
- Quality Criteria for Clustering (Coclustering)
- Measure of Variable Importance for a Clustering (Coclustering)
- Automatic tuning of Cluster Number (Cocluster Number)
- Clustering for Big Data
- Method to assess the evolution of a Clustering (Coclustering)
- Constrainted Clustering (Coclustering)
The list of application domain is includes, but it is not limited to:
- Evolving textual information analysis
- Evolving social network analysis
- Dynamic process control and tracking
- Intrusion and anomaly detection
- Genomics and DNA micro-array data analysis
- Adaptive recommender and filtering systems
A list of Applicative domains could be found in
P. Berkhin « Survey of clustering data mining techniques », Accrue Software, San Jose, CA, 2002.
Vincent Lemaire - Orange Labs
2 avenue Pierre Marzin, 2300 Lannion
Research Gate [...]
Pascal Cuxac, - INIST - CNRS
2 allée du Parc de Brabois, CS 10310, 54519 Vandœuvre les Nancy Cedex
Email : pascal.cuxac[at]inist.fr,
Jean-Charles Lamirel - LORIA – SYNALP Research Team
Campus Scientifique, BP. 239, 54506 Vandoeuvre les Nancy Cedex
Email : lamirel[at]loria.fr,