This is the public code base that can reproduce the found numbers that are presented in my MSc thesis at the VU. Made by beau fonville working as an internship thesis student for Hapsah.
| excels | ||
| helper_functions | ||
| ontology_guidance | ||
| results | ||
| .gitcommit | ||
| .gitignore | ||
| analyze_rule_dynamics.py | ||
| diagnostics_extreme_and_combined.py | ||
| evaluate_ruleset.py | ||
| final_compare_accuracies_no_vervuiling.py | ||
| oud_maak_verv_dataset.py | ||
| README.md | ||
README
Reference code for the ontology-guided RIPPER experiments. It runs entirely from
the bundled Excel dataset (excels/non_dummy_columns_df_less_classes.xlsx) — no
database connection is needed.
How to run
Activate a python env and install deps
python -m venv .venv
source ./.venv/Scripts/activate
pip install -r requirements.txt
Run all experiments
python run_all_experiments.py
(The Weka JRip baseline additionally needs a Java runtime; it is skipped
automatically if java/weka.jar are unavailable.)
Further info
./resultsholds all saved data behind the figures, including the learned rules.- the results that were used are in the subfolder
/pooled(pooled over the 5 folds).