Skip to main content
. 2017 Oct 18;31(4):451–463. doi: 10.1007/s10278-017-0029-8

Table 5.

Classification scenarios with no statistically significant difference identified on sensitivity and specificity (95% confidence intervals)

Number of features Classifier Dataset Sensitivity Specificity
48 RF Unbalanced (0.656, 0.744) (0.829, 0.880)
6 RF Unbalanced (0.615, 0.706) (0.833, 0.884)
5 KNN Balanced (0.630, 0.738) (0.739, 0.834)
3 KNN Unbalanced (0.617, 0.709) (0.839, 0.889)
2 J48 Balanced (0.637, 0.744) (0.774, 0.864)