bartMachine: Machine Learning with Bayesian Additive Regression Trees
This provides a more efficient and feature-rich tool for data analysts and researchers using BART in R, though it is incremental as it builds on an existing method.
The authors tackled the problem of implementing Bayesian additive regression trees (BART) in R by developing a new package called bartMachine, which introduces features like variable selection and missing data handling, and it is significantly faster and more scalable than existing implementations.
We present a new package in R implementing Bayesian additive regression trees (BART). The package introduces many new features for data analysis using BART such as variable selection, interaction detection, model diagnostic plots, incorporation of missing data and the ability to save trees for future prediction. It is significantly faster than the current R implementation, parallelized, and capable of handling both large sample sizes and high-dimensional data.