/Rule-Mining

Rule-Mining based classification: a benchmark study

Primary LanguagePythonBSD 3-Clause "New" or "Revised" LicenseBSD-3-Clause

Rule-Mining

This repository contains the code used for the experiments of the paper "Rule-Mining based classification: a benchmark study".

Launch the experiments

To launch the experiments do: bash rule_mining/experiments/jobs/file_name.py, from the Rule-Mining directory.

Visualization

To do the visualisation and get some statistics for the rules do:

  • pyhton rule_mining/visualisation/figures_rules.py,
  • python rule_mining/visualisation/rule_characteristics_pervar_perclass.py,
  • python rule_mining/visualisation/scores_var_by_level.py,
  • python rule_mining/visualisation/similarity_main_all.py,
  • python rule_mining/visualisation/similarity_main_l1.py, from the Rule-Mining directory.

Attribution

If you use this code, please cite:

  • [1] Luck, M., Pallet, N., & Damon, C. (2017). Rule-Mining based classification: a benchmark study. arXiv preprint arXiv:1706.10199.