The R Journal: article published in 2018, volume 10:2

Explanations of Model Predictions with live and breakDown Packages
Mateusz Staniak and Przemysław Biecek , The R Journal (2018) 10:2, pages 395-409.

Abstract 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.

Received: 2018-05-01; online 2018-12-11, supplementary material, (1.7 KiB)
CRAN packages: pdp, lime, caret, mlr, DALEX, iml, live, breakDown, archivist, xgboost, party, data.table, e1071, glmnet, randomForest
CRAN Task Views cited directly: MachineLearning
CRAN Task Views implied by cited CRAN packages: MachineLearning, Environmetrics, HighPerformanceComputing, Multivariate, Survival, Cluster, Distributions, Finance, MissingData, ModelDeployment, Psychometrics, ReproducibleResearch

This article and supplementary materials are licensed under a Creative Commons Attribution 4.0 International license.

@article{RJ-2018-072,
  author = {Mateusz Staniak and Przemysław Biecek},
  title = {{Explanations of Model Predictions with live and breakDown
          Packages}},
  year = {2018},
  journal = {{The R Journal}},
  doi = {10.32614/RJ-2018-072},
  url = {https://doi.org/10.32614/RJ-2018-072},
  pages = {395--409},
  volume = {10},
  number = {2}
}