8 Empirical results - adaa-polsl/RuleKit GitHub Wiki

Experiments were performed in stratified 10-fold cross validation scheme on 50 classification and 47 regression datasets. Following software packages were evalulated.

Classification:

• RuleKit,
• RIPPER,
• CN2,
• AQ,

Regression:

• RuleKit,
• M5Rules.

All algorithms were run with default parameters. Detailed results of all investigated algorithms on particular datasets can be found here.