Table 2 Two machine learning algorithms - Support Vector Machine (SVM) and Neural Network (NN) - were implemented through the function call MLearn (of R package MLInterfaces). SVM was applied with default parameters. NN was applied with three nodes in the hidden layer, with weight decay set to 0.01. Five-fold cross-validation was performed by generating a partition function (for cross-validation), where the partitions are approximately balanced with respect to the distribution of cancer tissue types (SLC or CA) in training and test sets in each fold. Confusion matrix from the classifier output was collected for calculation of misclassification rates

From: Sepsis-associated pathways segregate cancer groups