Skip to main content
. Author manuscript; available in PMC: 2019 Jan 1.
Published in final edited form as: Knowl Inf Syst. 2017 Nov 17;54(1):151–170. doi: 10.1007/s10115-017-1133-2

Fig. 2.

Fig. 2

calibration curves based on using 5 equal frequency bins when we use logistic regression, SVM, and naïve Bayes classification models for the binary classification task in the liver-disorder UCI dataset. Considering the frequency of observations in the first and the second bin, we notice the violation of the isotonicity assumption that is made by IsoReg in all the classification models.