Introduces a Python package for generating counterfactual explanations for tree-based algorithms

The importance of interpretability in machine learning models is growing as they’re increasingly applied in real‐world scenarios. Understanding how models make decisions advantages not only the model’s users but additionally those that are affected by the selections made by the model. Counterfactual explanations have been developed to address this issue, as they permit individuals to grasp how they might achieve a desirable consequence by perturbing their original data. Within the short term, counterfactual explanation possibly demonstrates actionable suggestions to those that are affected by a machine learning model decision. For instance, a one who was rejected for a loan application could know what could have done to be accepted this time and that will be useful to enhance on their next application.

Lucic et al. [1] proposed FOCUS, which is designed to generate optimal distance counterfactual explanations to the unique data for all of the instances in tree‐based machine learning models.

CFXplorer is a Python package that generates counterfactual explanations for a given model and data through the use of the FOCUS algorithm. This text introduces and showcases how CFXplorer could be used for generating counterfactual explanations.

GitHub repo: https://github.com/kyosek/CFXplorer

Documentation: https://cfxplorer.readthedocs.io/en/latest/?badge=latest

PyPI: https://pypi.org/project/CFXplorer/

1. FOCUS algorithm
2. CFXplorer examples
3. Limitations
4. Conclusion
5. References

This section briefly introduces the FOCUS algorithm.

The generation of counterfactual explanations is an issue that has been addressed by several existing methods. Wachter, Mittelstadt, and Russell [2] formulated this problem into an optimisation framework, nonetheless, this…