Multiple Classifier Systems: 5th International Workshop, MCS 2004, Cagliari, Italy, June 9-11, 2004. Proceedings

Ludmila I. Kuncheva (auth.), Fabio Roli, Josef Kittler, Terry Windeatt (eds.)3540221441, 9783540221449, 9783540259664

The fusion of di?erent information sourcesis a persistent and intriguing issue. It hasbeenaddressedforcenturiesinvariousdisciplines,includingpoliticalscience, probability and statistics, system reliability assessment, computer science, and distributed detection in communications. Early seminal work on fusion was c- ried out by pioneers such as Laplace and von Neumann. More recently, research activities in information fusion have focused on pattern recognition. During the 1990s,classi?erfusionschemes,especiallyattheso-calleddecision-level,emerged under a plethora of di?erent names in various scienti?c communities, including machine learning, neural networks, pattern recognition, and statistics. The d- ferent nomenclatures introduced by these communities re?ected their di?erent perspectives and cultural backgrounds as well as the absence of common forums and the poor dissemination of the most important results. In 1999, the ?rst workshop on multiple classi?er systems was organized with the main goal of creating a common international forum to promote the diss- ination of the results achieved in the diverse communities and the adoption of a common terminology, thus giving the di?erent perspectives and cultural ba- grounds some concrete added value. After ?ve meetings of this workshop, there is strong evidence that signi?cant steps have been made towards this goal. – searchers from these diverse communities successfully participated in the wo- shops, and world experts presented surveys of the state of the art from the perspectives of their communities to aid cross-fertilization.

---

Table of contents :
Front Matter….Pages –
Classifier Ensembles for Changing Environments….Pages 1-15
A Generic Sensor Fusion Problem: Classification and Function Estimation….Pages 16-30
AveBoost2: Boosting for Noisy Data….Pages 31-40
Bagging Decision Multi-trees….Pages 41-51
Learn++.MT: A New Approach to Incremental Learning….Pages 52-61
Beyond Boosting: Recursive ECOC Learning Machines….Pages 62-71
Exact Bagging with k -Nearest Neighbour Classifiers….Pages 72-81
Yet Another Method for Combining Classifiers Outputs: A Maximum Entropy Approach….Pages 82-91
Combining One-Class Classifiers to Classify Missing Data….Pages 92-101
Combining Kernel Information for Support Vector Classification….Pages 102-111
Combining Classifiers Using Dependency-Based Product Approximation with Bayes Error Rate….Pages 112-121
Combining Dissimilarity-Based One-Class Classifiers….Pages 122-133
A Modular System for the Classification of Time Series Data….Pages 134-143
A Probabilistic Model Using Information Theoretic Measures for Cluster Ensembles….Pages 144-153
Classifier Fusion Using Triangular Norms….Pages 154-163
Dynamic Integration of Regression Models….Pages 164-173
Dynamic Classifier Selection by Adaptive k-Nearest-Neighbourhood Rule….Pages 174-183
Spectral Measure for Multi-class Problems….Pages 184-193
The Relationship between Classifier Factorisation and Performance in Stochastic Vector Quantisation….Pages 194-203
A Method for Designing Cost-Sensitive ECOC….Pages 204-213
Building Graph-Based Classifier Ensembles by Random Node Selection….Pages 214-222
A Comparison of Ensemble Creation Techniques….Pages 223-232
Multiple Classifiers System for Reducing Influences of Atypical Observations….Pages 233-242
Sharing Training Patterns among Multiple Classifiers….Pages 243-252
First Experiments on Ensembles of Radial Basis Functions….Pages 253-262
Random Aggregated and Bagged Ensembles of SVMs: An Empirical Bias–Variance Analysis….Pages 263-272
Building Diverse Classifier Outputs to Evaluate the Behavior of Combination Methods: The Case of Two Classifiers….Pages 273-282
An Empirical Comparison of Hierarchical vs. Two-Level Approaches to Multiclass Problems….Pages 283-292
Experiments on Ensembles with Missing and Noisy Data….Pages 293-302
Induced Decision Fusion in Automated Sign Language Interpretation: Using ICA to Isolate the Underlying Components of Sign….Pages 303-313
Ensembles of Classifiers Derived from Multiple Prototypes and Their Application to Handwriting Recognition….Pages 314-323
Network Intrusion Detection by a Multi-stage Classification System….Pages 324-333
Application of Breiman’s Random Forest to Modeling Structure-Activity Relationships of Pharmaceutical Molecules….Pages 334-343
Experimental Study on Multiple LDA Classifier Combination for High Dimensional Data Classification….Pages 344-353
Physics-Based Decorrelation of Image Data for Decision Level Fusion in Face Verification….Pages 354-363
High Security Fingerprint Verification by Perceptron-Based Fusion of Multiple Matchers….Pages 364-373
Second Guessing a Commercial’Black Box’ Classifier by an’In House’ Classifier: Serial Classifier Combination in a Speech Recognition Application….Pages 374-383
Back Matter….Pages –