pmlbr and treeheatr: visualizing decision trees on benchmark datasets

The presentation of the decision tree with data represented as a heatmap is a new visualization that uncovers the tree’s performance, the data’s correlation structure, and the importance of each feature in predicting the outcome. Implemented in an easily installed package with a detailed vignette, treeheatr can be a useful teaching tool to enhance students’ understanding of a simple decision tree model before diving into more complex tree-based machine learning methods.

Trang is a postdoctoral fellow with Jason Moore at the Computational Genetics Lab, University of Pennsylvania. She enjoys developing machine learning methods for analyses of biomedical data, including neuroimage (functional/structural MRI), transcriptomics and genotypes. Most of the datasets she works with are high dimensional (i.e., have many predictors/features), so she spends most of her time building feature selection algorithms for these data. She trades her bias toward the nearest-neighbor concept for lower variance of her methods and better generalizability. When she is not knee deep in data, she runs, dance and seasonally skis.