MCS was organized by the Institute of Information Theory and Automation of the Academy of Sciences, Czech Republic, the University of Surrey, United Kingdom and the University of Cagliari, Italy.

MCS is a well-established series of meetings that provides an international forum for the discussion of issues in multiple classifier system design. The workshop achieved its objective to bring together 65 researchers from 22 countries of 4 continents, from diverse communities concerned with this topic, including neural network, pattern recognition, machine learning, and statistics. The Prague workshop was a successful continuation of six previous workshops organized in Italy, UK and USA.

This three day IAPR workshop was focused on the application of multiple classifier systems to biometrics.  This particular application area exercises all aspects of multiple classifier fusion, from intramodal classifier combination, through confidence-based fusion, to multimodal biometric systems.

MCS 2007’s technical programme was constituted by 49 regular papers, selected from more than 80 submissions. The presentations were grouped into nine sections: Kernel-Based Fusion, Boosting, Cluster and Graph Ensembles, Applications, Feature Subspace Ensembles, Multiple Classifier System Theory, Intramodal and Multimodal Fusion of Biometric Experts, Majority Voting,  and Ensemble Learning. An interesting panel discussion  entitled Multiple Classifier Systems - 10 years on: are we any wiser?  chaired by Lucy Kuncheva, University of Bangor, UK, helped to reflect on the current state of the art, progress made over the last decade, as well as to list open problems in the MCS field.  The “icing on the cake” was contributed by the three invited speakers: Dr Samy Bengio from Google, Professor Pramod Varshney from the Syracuse University, USA, and Professor Jon Benediktsson from the University of Iceland.

The MCS2007 workshop was co-sponsored by the International Association for Pattern Recognition and its Technical Committee TC1: Statistical Pattern Recognition, CSKI, and the European Union Network of Excellence in Biometrics Biosecure and in Multimedia Information Retrieval Muscle. The workshop proceedings were published in the Lecture Notes on Computer Science volume 4472 under the title Multiple Classifier Systems. The financial contribution of IAPR was used to fund four scholarships.

Workshop Report:  MCS 2007