Costs-Sensitive Classification in Two-Stage Binary Classifier
In the paper the problem of cost in the two-stage binary classifier is presented. Assuming that both the tree structure and the feature used at each non-terminal node have been specified, we present the expected total cost for two cases. The first one concerns the zero-one loss function, the second concerns the stage-dependent loss function. The work focuses on the difference between the expected total costs for these two cases of loss function. Obtained results relate to the globally optimal strategy of Bayes multistage classifier.
Unable to display preview. Download preview PDF.
- 4.Knoll, U., Nakhaeizadeh, G., Tausend, B.: Cost-sensitive pruning of decision trees. In: Bergadano, F., De Raedt, L. (eds.) ECML 1994. LNCS, vol. 784, pp. 383–386. Springer, Heidelberg (1994)Google Scholar
- 5.Núñez, M.: The use of background knowledge in decision tree induction. Machine Learning 6(3), 231–250 (1991)Google Scholar
- 6.Penar, W., Woźniak, M.: Experiments on classifiers obtained via decision tree induction methods with different attribute acquisition cost limit. AISC, vol. 45, pp. 371–377 (2007)Google Scholar
- 9.Tan, M.: Cost-sensitive learning of classification knowledge and its applications in robotics. Machine Learning 13, 7–33 (1993)Google Scholar
- 10.Turney, P.: Cost-sensitive classificcation: Empirical evaluation of a hybrid genetic decision tree induction algorithm. Journal of Artificial Intelligence Research 2, 369–409 (1995)Google Scholar