The Influence of Class Imbalance on Cost-Sensitive Learning: An Empirical Study
- 1 December 2006
- conference paper
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE International Conference on Data Mining (ICDM)
- No. 15504786,p. 970-974
- https://doi.org/10.1109/icdm.2006.158
Abstract
In real-world applications the number of examples in one class may overwhelm the other class, but the primary interest is usually on the minor class. Cost-sensitive learning has been deeded as a good solution to these class-imbalanced tasks, yet it is not clear how does the class-imbalance affect cost-sensitive classifiers. This paper presents an empirical study using 38 data sets, which discloses that class-imbalance often affects the performance of cost-sensitive classifiers: When the misclassification costs are not seriously unequal, cost-sensitive classifiers generally favor natural class distribution although it might be imbalanced; while when misclassification costs are seriously unequal, a balanced class distribution is more favorable.Keywords
This publication has 9 references indexed in Scilit:
- ON MULTI‐CLASS COST‐SENSITIVE LEARNINGComputational Intelligence, 2010
- Training cost-sensitive neural networks with methods addressing the class imbalance problemIEEE Transactions on Knowledge and Data Engineering, 2005
- Improving classifier utility by altering the misclassification cost ratioPublished by Association for Computing Machinery (ACM) ,2005
- Mining with rarityACM SIGKDD Explorations Newsletter, 2004
- EditorialACM SIGKDD Explorations Newsletter, 2004
- Cost-sensitive learning by cost-proportionate example weightingPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2004
- Learning When Training Data are Costly: The Effect of Class Distribution on Tree InductionJournal of Artificial Intelligence Research, 2003
- An instance-weighting method to induce cost-sensitive treesIEEE Transactions on Knowledge and Data Engineering, 2002
- MetaCostPublished by Association for Computing Machinery (ACM) ,1999