Area under the ROC curve by bubble-sort approach (BSA)

Authors:
Petr Honzík
Affiliations:
Department of Control and Instrumentation, Brno University of Technology, Faculty of Electrical Engineering and Communication, Brno, Czech Republic
Venue:
ACMOS'05 Proceedings of the 7th WSEAS international conference on Automatic control, modeling and simulation
Year:
2005

Citing 6
Cited 0

The Case against Accuracy Estimation for Comparing Induction Algorithms

ICML '98 Proceedings of the Fifteenth International Conference on Machine Learning
Tree induction vs. logistic regression: a learning-curve analysis

The Journal of Machine Learning Research
Optimising area under the ROC curve using gradient descent

ICML '04 Proceedings of the twenty-first international conference on Machine learning
On the Application of ROC Analysis to Predict Classification Performance Under Varying Class Distributions

Machine Learning
Regression Modeling Strategies

Regression Modeling Strategies
The use of the area under the ROC curve in the evaluation of machine learning algorithms

Pattern Recognition

Quantified Score

Hi-index	0.00

Visualization

Abstract

A new approach to area under ROC curve (AUC) evaluation is introduced and compared with the current methods. The main idea is based on the Bubble-Sort method. It advantages from the approach to the qualitative dependent variable which is used as the ordinal (and not nominal) variable in comparison to the classical approach. For binary output data the algorithm reaches the same complexity and values as the other methods. For multi-class classification the complexity differs from the classical approaches and the BSA values of AUC don't fail in the special cases as the current methods do.