Hierarchical cluster dendrogram.

<p>The algorithms used in the study are clustered into five clusters. Cluster 1 and 2 involve the algorithms (RLDA: Robust linear discriminant analysis, bagKNN: Bagged k-nearest neighbors, MDA: Mixture discriminant analysis, KNN: k-Nearest neighbors, SVMrbf: Support vector machines with radial basis function kernel, FDA: Flexible discriminant analysis, J48, C5.0, NN: Neural networks, SVMlin: Support vector machines with linear kernel, lsSVMrbf: Least squares support vector machines with radial basis function kernel, RF: Random forests, bagSVM: Bagged support vector machines), which are loaded on the positive side of the first principal component, and cluster 3 to 5 include the algorithms (LDA: Linear discriminant analysis, lsSVMlin: Least squares support vector machines with linear kernel, NSC: Nearest shrunken centroids, PLS: Partial least squares, QDA: Quadratic discriminant analysis, RQDA: Robust quadratic discriminant analysis, CIT: Conditional inference tree, NB: Naïve bayes, LVQ: Learning vector quantization, CART: Classification and regression trees) that are loaded on the negative side of the first principal component.