Skip to main content
. 2017 Feb 8;7:42192. doi: 10.1038/srep42192

Figure 1. Ideal rate and mean absolute error in train and test partitions for nine techniques averaged over 100 round of resampling process results for different models fitted.

Figure 1

Predicted dose within 20% of the actual dose in the train (A) and test (B) set of the derivation cohort. Mean absolute error between the predicted and actual dose in the train (C) and test (D) set. The vertical bars represent the 95% CIs of MAE. MLR: multiple linear regression; SVR: support vector regression; ANN: artificial neural network; RT: regression tree; RFR: random forest regression; BRT: boosted regression tree; MARS: multivariate adaptive regression splines; LAR: lasoo regression; BART: Bayesian additive regression trees.