Evaluating Explainable Methods for Predictive Process Analytics: A Functionally-Grounded Approach

The code used in this repository is adapted from the outcome-oriented predictive monitoring benchmark and a previous work on interpreting this benchmark. We thank the authors of the benchmark on outcome oriented predictions, for providing the source code which enabled further study on the interpretability of the models. In addition to this, a module developed to measure stability of feature subsets is used.

Name		Name	Last commit message	Last commit date
Latest commit History 6 Commits
.ipynb_checkpoints		.ipynb_checkpoints
Thesis Appendices		Thesis Appendices
bpic2012		bpic2012
bucketers		bucketers
figures		figures
params		params
preprocessing		preprocessing
production		production
sepsis_cases		sepsis_cases
transformers		transformers
.gitattributes		.gitattributes
.gitignore		.gitignore
BucketFactory.py		BucketFactory.py
DatasetManager.py		DatasetManager.py
DatasetManager_for_colab.py		DatasetManager_for_colab.py
EncoderFactory.py		EncoderFactory.py
README.md		README.md
dataset_confs.py		dataset_confs.py
dataset_confs_for_colab.py		dataset_confs_for_colab.py
generate_dataset_stats.ipynb		generate_dataset_stats.ipynb
generate_model.ipynb		generate_model.ipynb
optimise_params.ipynb		optimise_params.ipynb
optimize_params.py		optimize_params.py
test_fidelity.ipynb		test_fidelity.ipynb
test_stability.ipynb		test_stability.ipynb
test_timing.ipynb		test_timing.ipynb
update_dict.ipynb		update_dict.ipynb

Provide feedback