Reject option with multiple thresholds
暂无分享,去创建一个
can be obtained using the so-called “reject” option. Namely, the patterns that are the most likely to be misclassified are rejected (i.e., they are not classified); they are then handled by more sophisticated procedures (e.g., a manual classification is performed). However, handling high reject rates is usually too time-consuming for application purposes. Therefore, a trade-off between error and reject is mandatory. The formulation of the best error-reject trade-off and the related optimal reject rule was given by Chow [1]. According to Chow’s rule, a pattern x is rejected if: max P P T k N k i = ( ) = ( ) < 1, , | | K ω ω x x (3)
[1] Kagan Tumer,et al. Error Correlation and Error Reduction in Ensemble Classifiers , 1996, Connect. Sci..
[2] Mario Vento,et al. A method for improving classification reliability of multilayer perceptrons , 1995, IEEE Trans. Neural Networks.
[3] Fabio Roli,et al. Multisensor Image Recognition by Neural Networks with Understandable Behavior , 1996, Int. J. Pattern Recognit. Artif. Intell..
[4] Fabio Roli,et al. Multiple Reject Thresholds for Improving Classification Reliability , 2000, SSPR/SPR.