Opening Session : Alternatives to ROC Curve
09:00
Introduction
09:10
Invited Talk
Cost Curves
Rob Holte
10:10
The Expected Performance Curve.
Samy Bengio,
Johnny Mariéthoz, and Mikaela Keller.
10:30
COFFE BREAK
11:00
Boosting for Regression Using Regression Error Characteristic Curves.
Aloìsio
Carlos de Pina and Gerson Zaverucha.
11:20
Modifying ROC Curves to Incorporate Predicted Probabilities.
Cèsar
Ferri, Peter Flach, José Hernández-Orallo, and Athmane
Senad.
11:40
A scored AUC Metric for Classifier Evaluation and Selection
Shaomin Wu
and Peter Flach.
AUC Based Learning
12:00
Bagging Evolutionary ROC-based Hypotheses Application to Terminology
Extraction.
Jérôme Azé, Mathieu Roche, and Michèle Sebag.
12:20
Optimal Linear Combination of Dichotomizers via AUC.
Claudio
Marrocco, Mario Molinara and Francesco Tortorella.
12:40
LUNCH BREAK
14:30
AUC Maximizing Support Vector Learning.
Ulf Brefeld
and Tobias Scheffer.
Statistical
Analysis of ROC
14:45
A data-dependent generalisation error bound for the AUC.
Nicolas
Usunier, Massih-Reza Amini, and Patrick Gallinari.
15:05
Pointwise ROC Confidence Bounds: An Empirical Evaluation & ROC
Confidence Bands:
An Empirical Evaluation.
Sofus A.
Macskassy, Foster Provost and Saharon Rosset
15:40
BREAK
MultiClass ROC Analysis
16:10
Improving Classification Performance by Exploring the Role of Cost
Matrices in
Partioning the Estimated Class Probability Space.
Deirdre B. O'Brien
and Robert M. Gray.
16:30
Formulation and comparison of multi-class ROC surfaces.
Visualisation of multi-class ROC surfaces.
Jonathan E.
Fieldsend and Richard M. Everson.
17:05
Round Table + Wrap-up discussions
17:40
End