Fig. 4: Cross-validation techniques.
From: Systematic review of machine learning applications using nonoptical motion tracking in surgery
Cross-validation techniques presented as technique description, (plus sign +) advantages, and (minus sign -) disadvantages. Consists of hold out17,19,28,31,33,43,44,46,51,59,73,82,85, k-fold10,15,18,27,37,42,49,52,53,57,70,79,84,89,91, stratified k-fold29, leave-one-out21,35,37,49,62,64,72,87,88,90,92,93,95, leave one user out18,20,23,31,34,35,39,41,42,48,56,68,74,75,77,89,96, leave one trial out26,75,89, leave one super-trial out26,35,36,38,39,46,47,56,58.
