mars_domain_score: Calculate the MARS model domain applicability score

In viraldomain: Applicability Domain Methods of Viral Load and CD4 Lymphocytes

Calculate the MARS model domain applicability score

Description

This function fits a MARS (Multivariate Adaptive Regression Splines) model to the provided data and computes a domain applicability score based on PCA distances.

Usage

mars_domain_score(
  featured_col,
  train_data,
  mars_hyperparameters,
  test_data,
  threshold_value
)

Arguments

featured_col	The name of the featured column.
train_data	A data frame containing the training data.
mars_hyperparameters	A list of hyperparameters for the MARS model, including: num_terms: The number of terms to include in the MARS model. prod_degree: The degree of interaction terms to include. prune_method: The method used for pruning the MARS model.
test_data	A data frame containing the test data.
threshold_value	The threshold value for the domain score.

Value

A tibble with the domain applicability scores.

Examples

set.seed(123)
library(dplyr)
featured_col <- "cd_2022"
# Specifying features for training and testing procedures
train_data = viral %>%
  dplyr::select(cd_2022, vl_2022)
test_data = sero 
mars_hyperparameters <- list(num_terms = 3, prod_degree = 1, prune_method = "none")
threshold_value <- 0.99
# Call the function

viraldomain documentation built on June 8, 2025, 10:06 a.m.