Learn R Programming

RFmarkerDetector (version 1.0.1)

plotAUCvsCombinations: Plotting the average AUC as a function of the number of combinations

Description

This function allows to plot the average AUC as a function of the number k-combinations of the n input variables. If n is the number of input variables, the number of k-combinations of those variables is equal to $n!/k!(n!-k!)$. Each of these combinations contains the indexes of the input variables selected. For each combination we can extract a dataset, build a random forest model and perform a cross-validation. We can describe the performance of each cross-validated model with an 'average' ROC curve and its AUC. The collected auc values for each combination (dataset) are used by the function to build a diagram of the AUC as a function of the number of combinations

Usage

plotAUCvsCombinations(auc_values, num_of_variables, num_of_combinations)

Arguments

auc_values
an array with the auc values to plot
num_of_variables
the k dimension of each combination
num_of_combinations
the number of k combinations of the set of the input variables