Multiple classifier systems

the long version

Multiple Classifier Systems: 5th International Workshop, MCS 2004, Cagliari, Italy, June 9-11, 2004. Proceedings<br />Author: Fabio Roli, Josef Kittler, Terry Windeatt<br /> Published by Springer Berlin Heidelberg<br /> ISBN: 978-3-540-22144-9<br /> DOI: 10.1007/b98227<br /><br />Table of Contents:<p></p><ul><li>Classifier Ensembles for Changing Environments </li><li>A Generic Sensor Fusion Problem: Classification and Function Estimation </li><li>AveBoost2: Boosting for Noisy Data </li><li>Bagging Decision Multi-trees </li><li>Learn++.MT: A New Approach to Incremental Learning </li><li>Beyond Boosting: Recursive ECOC Learning Machines </li><li>Exact Bagging with k-Nearest Neighbour Classifiers </li><li>Yet Another Method for Combining Classifiers Outputs: A Maximum Entropy Approach </li><li>Combining One-Class Classifiers to Classify Missing Data </li><li>Combining Kernel Information for Support Vector Classification </li><li>Combining Classifiers Using Dependency-Based Product Approximation with Bayes Error Rate </li><li>Combining Dissimilarity-Based One-Class Classifiers </li><li>A Modular System for the Classification of Time Series Data </li><li>A Probabilistic Model Using Information Theoretic Measures for Cluster Ensembles </li><li>Classifier Fusion Using Triangular Norms </li><li>Dynamic Integration of Regression Models </li><li>Dynamic Classifier Selection by Adaptive k-Nearest-Neighbourhood Rule </li><li>Spectral Measure for Multi-class Problems </li><li>The Relationship between Classifier Factorisation and Performance in Stochastic Vector Quantisation </li><li>A Method for Designing Cost-Sensitive ECOC</li></ul>