Technological developments of computers technology and its recent applications provide access to new types of data, which have not been considered in classical machine learning and pattern recognition algorithms. Two particularly interesting characteristics of such data sets include their huge size and streaming nature. Big data sets are now collected in many fields as, e.g., finance, business, medical systems, internet, biology and many other scientific research. Such data exceeds the processing capacity of conventional systems but there is a need for their real-time analysis and discovering unknown patterns. As standard methods cannot handle the complex structure and size of massive data, fulfill computational cost-effective requirements, these topics receive a growing interest both from academia and commercial companies. Moreover, data sets rapidly increase their size as they are often generated in a form of incoming streams. Sensor networks, monitoring, traffic management, tele...
Read more
Technological developments of computers technology and its recent applications provide access to new types of data, which have not been considered in classical machine learning and pattern recognition algorithms. Two particularly interesting characteristics of such data sets include their huge size and streaming nature. Big data sets are now collected in many fields as, e.g., finance, business, medical systems, internet, biology and many other scientific research. Such data exceeds the processing capacity of conventional systems but there is a need for their real-time analysis and discovering unknown patterns. As standard methods cannot handle the complex structure and size of massive data, fulfill computational cost-effective requirements, these topics receive a growing interest both from academia and commercial companies. Moreover, data sets rapidly increase their size as they are often generated in a form of incoming streams. Sensor networks, monitoring, traffic management, telecommunication, or web log analysis are examples of such applications where machines working in dynamic environments continuously generate data. Compared to static environments the processing of data streams implies new requirements for algorithms, such as constraints on memory usage, restricted learning and testing time, and one scan of incoming instances. Another important aspect of learning classifiers from streams refers to changes in the data distributions and target concepts over time. Detecting theses changes and adapting classifiers to concept drifts becomes one of the challenges for new learning algorithms. Furthermore learning classifiers and other predictive models from massive and stream data intersects with other related problems as handling class distributions, detecting rare cases or semi-supervised learning.
<br>
<br>To address the research topics concerning both data stream classification and massive data analytics issues we organize a Special Session co-located within the framework of the 9th Computer Recognition Systems (CORES) 2015 Conference, which will be held from 25 to 27 May 2015 in Wroclaw, Poland. The main aim of this session is to gather researchers interested in these issues and to demonstrate results in these and related areas of mining such difficult data. Original and high-quality novel research dedicated to both theory and applications of big and stream data processing algorithms are invited.
<br>
<br>IMPORTANT: DSBD2015 is part of the CORES2015 conference. 80 top quality papers accepted for the CORES2015 will have the entire registrasion fee reimbused thanks to the support from European Union and the ENGINE project:
<br>
<br>http://engine.pwr.wroc.pl/en/
<br>
<br>
<br>Scope of the Special Session
<br>
<br>The topics of this Special Session are concentrated on (but not limited to) following themes:
<br>
<br> Incremental online learning algorithms
<br> Detection and adaptation to the presence of the concept drift
<br> One-class classification and novelty detection
<br> Imbalanced classification
<br> Discovery, detection and classification of complex patterns in massive or evolving data
<br> Semi-supervised approaches and active learning paradigms
<br> Scaling-up learning algorithms
<br> Near real-time analysis of massive data
<br> Integration and fusion of heterogeneous data structures and streams
<br> Distributed and parallel computing systems for big data analytics
<br> Problem of privacy in big and stream data
<br> Applications to real-life problems from medicine, bioinformatics, multimedia, sensors, social networks and related domains
<br>
<br>Special Session Organisers
<br>
<br>Bartosz Krawczyk, Wrocław University of Technology, bartosz.krawczyk@pwr.edu.pl
<br>
<br>Jerzy Stefanowski, Poznań University of Technology, jerzy.stefanowski@cs.put.poznan.pl
<br>
<br>Michał Woźniak, Wrocław University of Technology, michal.wozniak@pwr.edu.pl
<br>
<br>Important dates
<br>Submission of papers: March 1, 2015
<br>Notification of acceptance: April 15, 2015
<br>Submission of camera-ready papers: May 1, 2015
<br>CORES conference: May 25-27, 2015
<br>Paper submission instructions
<br>
<br>All accepted papers will be published by Springer in the Advances in Intelligent and Soft Computing series volume only in the case if the paper is presented during the session. There is a 10 page limit. Please prepare your article according to http://www.springer.com/series/4240.
<br>
<br>In case of any questions please contact directly the Special Session Organisers.
