3.1.8 Cost Curve
Jiang et al. [19] recommended adopting cost curves for the fault prediction
performance evaluation. This is the first study to propose cost curves for
performance evaluation of fault predictors and it is not yet widely used. However,
it is not easy to determine the misclassification cost ratio and the selection of this
parameter can make the model debatable. Drummond and Holte [10] proposed
cost curves to visualize classifier performance and the cost of misclassification
was included in this technique. Cost curve plots the probability cost function on
the x-axis and the normalized expected misclassification cost on the y-axis.
Details of this approach will not be given here due to length considerations, and