Skip to main content
. 2022 Mar 3;77:103911. doi: 10.1016/j.ebiom.2022.103911

Figure 1.

Fig 1

Heatmaps illustrating the performance of each machine learning algorithm (rows) with each feature reduction method (columns), measured by validation set AUC. No FR: No feature reduction (full feature set used), LASSO: Least Absolute Shrinkage and Selection Operator, E Net: Elastic-Net, RFE: Recursive Feature Elimination, Univariate LR: Univariate Logistic Regression, XGB: Extreme Gradient Boosting machine, NB: Naïve-Bayes, PSL: Partial Least Squares, L-SVM: Linear Support Vector Machine, NL-SVM: Non-linear (radial) SVM, RF: Random Forest, MDA: Mixture Discriminant Analysis, KNN: K-Nearest Neighbours, GLM: Generalised Linear Model, NNET: Neural Network.