|
Workshop Report: MCS 2009 |
|
Report prepared by: Carlo Sansone (Italy) |
|
Workshop Chairs:
Jón Atli Benediktsson (Iceland) Josef Kittler. IAPR Fellow(UK) Fabio Roli, IAPR Fellow (Italy) |
|
MCS 2009 was organized by the Faculty of Electrical and Computer Engineering of the University of Iceland in association with the Center for Vision, Speech and Signal Processing of the University of Surrey, UK and the Department of Electrical and Electronic Engineering of the University of Cagliari, Italy. The workshop was co-sponsored by the International Association for Pattern Recognition and its Technical Committee TC1: Statistical Techniques in Pattern Recognition, and by the IEEE Geoscience and Remote Sensing Society and the IEEE Iceland Section. MCS is a well-established series of IAPR meetings that provides an international forum for the discussion of issues in multiple classifier systems' theory and applications. The Reykjavik workshop was in fact the successful continuation of seven previous workshops organized in Italy (Cagliari, three times), UK (Cambridge and Guildford), USA (Seaside, CA) and Czech Republic (Prague). The 2009 edition achieved its objective to bring together 63 researchers from 17 countries on all five continents, from diverse communities concerned with multiple classifier systems, including pattern recognition, machine learning and statistics. From the applicative point of view, the focus of MCS 2009 was on remote sensing. This particular application uses multiple classifiers for raw data fusion, feature level fusion and decision level fusion. Outstanding contributions were made by the invited speakers, Melba Crawford from Purdue University and Zhi-Hua Zhou of Nanjing University. Prof. Crawford’s expertise in the use of multiple classification systems in remote sensing made the discussions on this topic at MCS 2009 particularly fruitful. Zhi-Hua Zhou also gave a very interesting talk that highlighted the reasons why semi-supervised and ensemble learning can benefit each other. Apart from the invited talks, MCS 2009’s technical program was constituted by 54 regular papers, selected from more than 70 submissions. The presentations were grouped into a poster session, on MCS Methods and Applications, and ten oral sessions: ECOC, Boosting and Bagging; MCS in Remote Sensing; Unbalanced Data and Decision Templates; Stacked Generalization and Active Learning; Concept Drift, Missing Values and Random Forest; SVM Ensembles; Fusion of Graphs, Concepts and Categorical Data; Clustering; Classifier and Feature Selection; and Theory of MCS. Coming to the social program, the gala dinner was held in a lovely Icelandic restaurant in Videy, a small island near Reykjavik. The excellent quality of the food provided by the Videy House (the first building constructed of stone in Iceland!) together with the performance provided by a group of Icelandic artists was a highlight of this high-quality scientific workshop. Finally, an announcement for the whole MCS community. Starting from the next edition, MCS comes back to its initial annual format. The 2010 edition will take place in Cairo, Egypt, organized by the Nile University. |
|
Proceedings of the conference have been published by Springer-Verlag in the Lecture Notes in Computer Science Series (Volume 5519) titled Multiple Classifier Systems.
Click on the image to go to the publisher’s web site for this volume. |
|
MCS2009 chair, Jón Atli Benediktsson, during the opening session |
