Complex models are commonly used in predictive modeling. In this paper we present R packages that can be used for explaining predictions from complex black box models and attributing parts of these predictions to input features. We introduce two new approaches and corresponding packages for such attribution, namely live and breakDown. We also compare their results with existing implementations of state-of-the-art solutions, namely, lime (Pedersen and Benesty, 2018) which implements Locally Interpretable Model-agnostic Explanations and iml (Molnar et al., 2018) which implements Shapley values.
Supplementary materials are available in addition to this article. It can be downloaded at RJ-2018-072.zip
pdp, lime, caret, mlr, DALEX, iml, live, breakDown, archivist, xgboost, party, data.table, e1071, glmnet, randomForest
MachineLearning, Environmetrics, HighPerformanceComputing, Multivariate, Survival, Cluster, Distributions, Finance, MissingData, ModelDeployment, Psychometrics, ReproducibleResearch
Text and figures are licensed under Creative Commons Attribution CC BY 4.0. The figures that have been reused from other sources don't fall under this license and can be recognized by a note in their caption: "Figure from ...".
For attribution, please cite this work as
Staniak & Biecek, "The R Journal: Explanations of Model Predictions with live and breakDown Packages ", The R Journal, 2018
BibTeX citation
@article{RJ-2018-072,
author = {Staniak, Mateusz and Biecek, Przemysław},
title = {The R Journal: Explanations of Model Predictions with live and breakDown Packages },
journal = {The R Journal},
year = {2018},
note = {https://doi.org/10.32614/RJ-2018-072},
doi = {10.32614/RJ-2018-072},
volume = {10},
issue = {2},
issn = {2073-4859},
pages = {395-409}
}