# --------------------------------------------
# CITATION file created with {cffr} R package
# See also: https://docs.ropensci.org/cffr/
# --------------------------------------------
 
cff-version: 1.2.0
message: 'To cite package "tree.interpreter" in publications use:'
type: software
license: MIT
title: 'tree.interpreter: Random Forest Prediction Decomposition and Feature Importance
  Measure'
version: 0.1.1
identifiers:
- type: doi
  value: 10.32614/CRAN.package.tree.interpreter
abstract: An R re-implementation of the 'treeinterpreter' package on PyPI <https://pypi.org/project/treeinterpreter/>.
  Each prediction can be decomposed as 'prediction = bias + feature_1_contribution
  + ... + feature_n_contribution'. This decomposition is then used to calculate the
  Mean Decrease Impurity (MDI) and Mean Decrease Impurity using out-of-bag samples
  (MDI-oob) feature importance measures based on the work of Li et al. (2019) <arXiv:1906.10845>.
authors:
- family-names: Sun
  given-names: Qingyao
  email: sunqingyao19970825@gmail.com
preferred-citation:
  type: article
  title: A Debiased MDI Feature Importance Measure for Random Forests
  authors:
  - family-names: Li
    given-names: Xiao
  - family-names: Wang
    given-names: Yu
  - family-names: Basu
    given-names: Sumanta
  - family-names: Kumbier
    given-names: Karl
  - family-names: Yu
    given-names: Bin
  url: http://arxiv.org/abs/1906.10845
  abstract: Tree ensembles such as Random Forests have achieved impressive empirical
    success across a wide variety of applications. To understand how these models
    make predictions, people routinely turn to feature importance measures calculated
    from tree ensembles. It has long been known that Mean Decrease Impurity (MDI),
    one of the most widely used measures of feature importance, incorrectly assigns
    high importance to noisy features, leading to systematic bias in feature selection.
    In this paper, we address the feature selection bias of MDI from both theoretical
    and methodological perspectives. Based on the original definition of MDI by Breiman
    et al. for a single tree, we derive a tight non-asymptotic bound on the expected
    bias of MDI importance of noisy features, showing that deep trees have higher
    (expected) feature selection bias than shallow ones. However, it is not clear
    how to reduce the bias of MDI using its existing analytical expression. We derive
    a new analytical expression for MDI, and based on this new expression, we are
    able to propose a debiased MDI feature importance measure using out-of-bag samples,
    called MDI-oob. For both the simulated data and a genomic ChIP dataset, MDI-oob
    achieves state-of-the-art performance in feature selection from Random Forests
    for both deep and shallow trees.
  date-accessed: '2019-10-18'
  journal: arXiv:1906.10845 [cs, stat]
  month: '6'
  year: '2019'
  notes: 'arXiv: 1906.10845'
  keywords:
  - Statistics - Machine Learning
  - Computer Science - Machine Learning
repository: https://nalzok.r-universe.dev
repository-code: https://github.com/nalzok/tree.interpreter
commit: 0a04a7a790aa128141fc3592ac70da20d90633d4
url: https://github.com/nalzok/tree.interpreter
date-released: '2020-01-28'
contact:
- family-names: Sun
  given-names: Qingyao
  email: sunqingyao19970825@gmail.com