This project was partially funded via the Omar N. Bradley Officer Research Fellowship in Mathematics.
This R package is for Binary Local Expert Regression, which is a proprietary ensemble method for regression tasks. Briefly, it involves decomposing the task of predicting a continuous target variable into many simpler probabilistic class estimation tasks. These 'local expert' models are reconstructed to form a unique distributional understanding of each separate instance, and a separate regression model is trained on top of this model output (see David Wolpert's work on stacked generalization).
This vignette walks through an application of Binary Local Expert Regression to chemical research data from Max Kuhn's Applied Predictive Modeling package, and serves as an introduction to the methodology.
If you have the devtools
package installed in R
you can download the (unstable) development version like this:
devtools::install_github('nnormandin/localexpeRt')
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.