Experiment repository for DMKD submission "Guiding the generation of counterfactual explanations through temporal background knowledge for Predictive Process Monitoring"

To set up the required packages run: pip install -e . This should install all required dependencies.

To run the experiments, datasets used in the experiments should be downloaded from: https://drive.google.com/file/d/1Jr4RM4dRlbi-aXAxNaifRVGLDOnhlIdN/view?usp=sharing Afterwards, unzip in the folder.

To run the experiments, just launch the run_simple_pipeline.py file.

The output will be in the experiments folder, including the generated counterfactuals, the evaluation results and the discovered DECLARE models.