Fostering interpretability of data mining models through data perturbation

作者：

Highlights：

• Black-box decision-making algorithms proposals can be counter-intuitive.

• Existing interpretability tools only extract global statistics.

• Focus on interpretability for singular cases needed for data-driven decision making.

• The algorithm is easy to implement, efficient and model-agnostic.

摘要

•Black-box decision-making algorithms proposals can be counter-intuitive.•Existing interpretability tools only extract global statistics.•Focus on interpretability for singular cases needed for data-driven decision making.•The algorithm is easy to implement, efficient and model-agnostic.

论文关键词：Interpretability,Data mining,Random forest,Artificial neural networks

论文评审过程：Received 19 February 2019, Revised 1 July 2019, Accepted 1 July 2019, Available online 2 July 2019, Version of Record 8 July 2019.

论文官网地址：https://doi.org/10.1016/j.eswa.2019.07.001