mrandri19/X-PLAIN

X-PLAIN: A Rule-based, model agnostic, classifier explainability framework

XPLAIN is a rule-based, model-agnostic method that explains the prediction of any classifier on a specific instance by analyzing the joint effect of feature subsets on the classifier prediction. The relevant subsets are identified by learning a local rule-based model in the neighborhood of the prediction to explain.

Installation

git clone https://github.com/mrandri19/Internship-Project-X-PLAIN.git
cd Internship-Project-X-PLAIN
pip install .

Screenshots

TODO

Tutorials (Sample Notebooks)

• Basic usage, comparison of different classifiers on the monks dataset

Papers

• Bring Your Own Data to X-PLAIN: https://dl.acm.org/doi/abs/10.1145/3318464.3384710

Publication:SIGMOD '20: Proceedings of the 2020 ACM SIGMOD International Conference on Management of DataJune 2020 Pages 2805–2808https://doi.org/10.1145/3318464.3384710