Workshop Schedule


Time
Presentation
9:00 - 9:25 William Klement, Nathalie Japkowicz, Chris Drummond
"Evaluation in Machine Learning: High Level Questions"

9:25 - 9:50 Janez Demsar
"On the Appropriateness of Statistical Tests in Machine Learning"
(pdf)

9:50 - 10:15 Edith Law
"The Problem of Accuracy as an Evaluation Criterion"
(pdf)

10:15 - 10:30 Discussion


10:30 - 11:00

Break
11:00 - 11:25 Niklas Lavesson, Paul Davidsson
"Towards Application-specific Evaluation Metrics"
(pdf)

11:25 - 11:50 William Klement, Peter Flach
"Soft Receiver Operating Characteristics Curves"
(pdf)

11:50 - 12:15 Mohak Shah
"Risk Bounds for Classifier Evaluation: Possibilities and Challenges"
(pdf)

12:15 - 12:30 Discussion

12:30 - 2:30

Lunch
2:30 - 2:55 Rita Ribeiro, Luis Torgo
"Utility-based Performance Measures for Regression"
(pdf)

2:55 - 3:20 Marina Sokolova, Khaled El Emam
"Evaluation of Learning from Screened Positive Examples"
(pdf)

3:20 - 3:45 Joao Gama, Pedro Pereira Rodrigues, Gladys Castillo
"Evaluating Algorithms that Learn from Data Streams"
(pdf)

3:45 - 4:00 Discussion

4:00 - 4:30

Break
4:30 - 4:55 Chris Drummond
"Finding a Balance between Anarchy and Orthodoxy"
(pdf)

4:55 - 5:20 Nathalie Japkowicz
"Classifier Evaluation: A Need for Better Education and Restructuring"
(pdf)

5:20 - 6:00
Panel Discussion