Home
Rule-based models are often used for data analysis as they combine interpretability with predictive power. We present RuleKit2, a versatile tool for rule learning. Based on a sequential covering induction algorithm, it comes with a set of useful features:
- Suitability for different kind of problems:
- classification,
- regression,
- survival.
- User-guided induction for verifying hypotheses concerning data dependencies which are expected or of interest.
- Contrast set mining.
- Different experimental environments:
- standalone command line tool implemented in Java,
- Python package: https://github.com/adaa-polsl/RuleKit-python,
- browser application with a graphical user interface: https://ruleminer.github.io/rulekit-gui.
- Documented Java API: https://adaa-polsl.github.io/RuleKit.
Unlike the first revision, RuleKit2 does not depend on RapidMiner. Running RuleKit as a RapidMiner plugin and R package is no longer supported in version 2. The overview of RuleKit2 architecture is presented below.
RuleKit provides latest versions of our algorithms (some of them were initially published as independent packages and integrated later):
- LR-Rules (Wróbel et al, 2017) - survival rules induction,
- GuideR (Sikora et al, 2019) - user-guided induction.
- RuleKit-CS (Gudyś et al, 2024) - contrast set mining.
The software requires Java Development Kit in version 8 to work properly. In Windows one can download the installer from Oracle webpage. In Linux, a system package manager should be used instead. For instance, in Ubuntu 16.04 execute the following command:
sudo apt-get install default-jdk