MDIL 2013 - special session on Multiple-Data and Intensive Learning

Dear Colleague,
We would like to cordially invite you to submit a paper for the special session on
Multiple-Data and Intensive Learning
Special Session at IJCNN 2013
https://sites.google.com/site/mdimijcnn2013/
Fairmont Hotel ? Dallas, TX
August 4?9, 2013
http://www.ijcnn2013.org/
organized within International Joint Conference on Neural Networks sponsored jointly by INNS, the IEEE Computational Intelligence Society.
In many fields, such as multimedia, geographical information systems, databases, bio-informatics, and with advances in data collection and storage technologies have allowed organizations to accumulate vast amounts of data (Terabyte, Petabyte, and sometimes Zettabyte). Often the data come from multiple sources such as sensors, data stream, web and have different distributions. Often, these data are stored and presented in such a way that they are not supported by traditional machine learning algorithms. Most existing learning algorithms are not suitable for massive data (Big Data).
Most machine learning methods operate a small proportion of information collected in their information system. Therefore, there is a growing need to run learning algorithms on very large and distributed datasets. Recent trends in hardware have brought new challenges to the programming community; multi-core and many-core systems have become ubiquitous. Data's explosion involves that machine learning algorithms be parallelized and distributed using the new parallelism paradigm. While the need for machine learning is evident, few solutions exist today. Some learning algorithms have migrated to the MapReduce programming model, that was designed to simplify the processing of large data set. While MapReduce paradigms become very popular, we believe they have a set of limitations making them ill-suited to the implementation of parallel and distributed machine learning. In order to discover knowledge it is necessary to explore data efficiently by utilizing enormous computing power and by using new paradigms of parallel and/or distributed processing.
We invite submissions of papers describing original research addressing multiple-data and intensive mining challenges in association with parallelism or distributed model.
This includes but is not restricted to the following topics:
Machine learning algorithms for Big Data
Distributed clustering, biclustering
Distributed ensemble classifier
Distributed neural networks
Parallel and distributed computational intelligence
Multiple view classifier/ clustering
Supercomputing for big data mining
Clustering, classification and fusion of multi-source, multimode and/or distributed data
Collaborative learning
Learning with multiple imperfect (imprecise, uncertain, incomplete) labelers and/or data
Data privacy
Crowd mining
Data mining with cloud computing
Compressive sampling, very high dimensional data analysis
Future research challenges of data intensive mining
**Innovative intensive data mining approaches are welcome.
Submission procedures: http://www.ijcnn2013.org/paper-submission.php#cont...
Further information: http://www.ijcnn2013.org
Important Dates:
Paper Submission Deadline February 22, 2013
Camera-Ready Paper Submission May 1, 2013
Organizers
Guillaume Cleuziou, University of Orleans, LIFO,
Cyril de Runz, University of Reims Champagne-Ardenne, CReSTIC,
Mustapha Lebbah, University of Paris 13, LIPN,
Cédric Wemmert, University of Strasbourg, LSIIT,
The session is supported by french group in "complex data mining" of EGC society.
Contact
mustapha.lebbah-AT-univ-paris13.fr, cyril.de-runz-AT-univ-reims.fr
Yours sincerely,
Guillaume, Cyril, Mustapha, and Cédric
--
------------------------------------------------------------------
Université Paris 13, Sorbonne Paris Cité,
Laboratoire d'Informatique de Paris-Nord (LIPN),
CNRS(UMR 7030),
99, av. J-B Clément
F-93430, Villetaneuse, France.
Tel: +331 49 40 38 94
Fax: +331 48 26 07 12
http://www-lipn.univ-paris13.fr/~lebbah