Evaluate Feature Selection Methods Using Supervised Learning • EvaluateFeatureSelection

Generates plots to visualize and assess the performance of feature selection methods using supervised learning. It also provides functions to plot scree plots to visualize good cutting points for the number of features to be selected.

Installation

You can install the development version of EvaluateFeatureSelection like so:

install.packages("remotes")
remotes::install_github("Ahmad-Alsaleh/EvaluateFeatureSelection")

Example

Generate a scree plot

library(EvaluateFeatureSelection)
features_scores <- c(x1 = 0.8165005, x2 = -0.1178857, ...)
get_scree_plot(features_scores)

Similarly, you can use get_auc_plot(...) or get_acc_plot(...) to evaluate the performance of feature selection methods using supervised learning and AUC/accuracy as the performance metric.