Table 4.

Performance on the test set of the three machine learning techniques under consideration.

Technique	Sensitivity, mean (95% CI)	PPVa, mean (95% CI)	NPVb, mean (95% CI)	Specificity, mean (95% CI)	F score, mean (95% CI)
GLMNetc	72.3 (66.4-78.1)	24.5 (21.0-28.0)	98.3 (97.9-98.6)	87.4 (85.4-89.5)	36.5 (32.2-40.8)
MAXENTd	74.8 (62.2-87.5)	11.0 (9.5-12.5)	98.0 (97.3-98.6)	65.5 (54.7-76.2)	19.1 (17.2-20.9)
Boosting	79.2 (69.7-88.7)	63.1 (42.7-83.5)	98.8 (98.3-99.3)	96.9 (94.2-99.6)	68.5 (59.3-77.7)

^aPPV: positive predicative value.

^bNPV: negative predicative value.

^cGLMNet: elastic-net regularized generalized linear model.

^dMAXENT: maximum entropy.