ROC and Cost Graphs for General Cost Matrix Where Correct Classifications Incur Non-zero Costs

- Journal title : Communications for Statistical Applications and Methods
- Volume 11, Issue 1, 2004, pp.21-30
- Publisher : The Korean Statistical Society
- DOI : 10.5351/CKSS.2004.11.1.021

Title & Authors

ROC and Cost Graphs for General Cost Matrix Where Correct Classifications Incur Non-zero Costs

Kim, Ji-Hyun;

Kim, Ji-Hyun;

Abstract

Often the accuracy is not adequate as a performance measure of classifiers when costs are different for different prediction errors. ROC and cost graphs can be used in such case to compare and identify cost-sensitive classifiers. We extend ROC and cost graphs so that they can be used when more general cost matrix is given, where not only misclassifications but correct classifications also incur penalties.

Keywords

cost-sensitive learning;cost matrix;iso-performance line;CART;

Language

English

Cited by

1.

The Unified Framework for AUC Maximizer,;;;;;

2.

비용곡선과 ROC곡선에서의 비용비율,홍종선;유현상;

References

1.

UCI Repository of machine iearning databases, 1998.

2.

Classification and Regression Trees, 1984.

3.

Technical Report, School of Information Technology and Engineering, University of Ottawa, 2000.

4.

Series in Cognition and Perception, 1975.

5.

Proceedings of the Seventeenth International Joint Conference on Artificial Intelligence, 2001.

6.

Proceedings of the Third International Conference on Knowledge Discovery and Data Mining, 1997.
pp.43-48

7.

Technical Report, Mayo Foundation, 1997.