Skip to main content

Table 2 The χ2 p-values for the fit to the diagonal in the reliability diagram, number of calibrated points, and difference between the maximum and minimum calibrated probabilities (range) for the SVM classifier presented in Fig. 3

From: A nonparametric Bayesian method of translating machine learning scores to probabilities in clinical decision support

Data set BBQ Proposed method
χ2 p-value Calibrated points Range χ2 p-value Calibrated Points Range
Lung Cancer <0.001 2 0.82 0.001 4 0.90
SPECT <0.001 5 0.75 <0.001 7 0.92
Parkinsons 0.01 8 1.0 0.651 6 0.95
Arcene 0.387 9 0.96 0.841 8 0.94
Suicide 0.048 9 0.94 0.013 8 0.90
Arrhythmia 0.521 5 0.66 0.001 9 0.87
Breast Cancer 0.003 8 1.0 0.001 7 1.0
Contraception 0.018 5 0.48 0.124 8 0.81
  1. The (Contraception) data set with a large overlap in the score distributions is emphasized in boldface. When compared with the other data sets, the proposed method produces a larger number of calibrated points, indicating a finer granularity in the calibrated probabilities