CPBG 2015 - 1st IEEE International Workshop on Classification Problems Embedded in the Nature of Big Data

One of the main challenges for the machine learning algorithms are problems arising from the nature of the data itself. Such problems as high dimensionality of the feature space, small availability of specific training patterns, class and feature noise, imbalanced class distribution etc. have a major negative impact on the overall predictive accuracy of pattern recognition systems. This is especially vivid in case of big data analytics, where these problems are further deepened by 3Vs (Variety, Velocity and Volume). Big data can be affected by a much higher noise or imbalance ratio than any standard dataset analysed so far in these domains. Additionally, with big data new problems arise such as analysing complex mutual dependencies among objects, high number of classes and non-stationary nature of incoming samples.
Therefore introducing new methodologies, ranging from separate data pre-processing methods to approaches embedded in the classifiers is of primary interest to the field. The aim of this workshop is to provide a forum to exchange new theoretical ideas and practical implementations in this field.
Scope and Interests
Topics of interest include, but are not limited to:
Addressing difficulties arising in analysis of big data, that are embedded in the nature of examples
New methods for imbalanced classification (pre-processing and classifier level)
Approaches for handling noise present in the data (label, feature or mixed noise)
Efficient learning techniques for classification of massive and highly dimensional data
Machine learning algorithms for tackling dynamic and evolving structures of data
One-class classification and novelty detection for big data
Semi-supervised and unsupervised classification over large datasets
Strategies for dealing with a high number of classes (multi-class and multi-label approaches) in massive collections of objects
Methods and architectures for fast and efficient processing of algorithms for handling difficult data
Applications of mentioned topics in medicine, engineering, finance and social media
Submission Instructions
Papers submitted to the workshop should be written in English conforming to the IEEE Conference Proceedings Format (8.5" x 11", Two-Column). The paper should be submitted through the workshop submission system at the workshop website. The length of the papers should not exceed 6 pages + 2 pages for over length charges.