ROCCER: An Algorithm for Rule Learning Based on ROC Analysis

Prati, Ronaldo C.; Flach, Peter A.

ROCCER: An Algorithm for Rule Learning Based on ROC Analysis

Download BibTeX Cite

Proceedings of the Nineteenth International Joint Conference on Artificial Intelligence. Pages 823-827.

Abstract

We introduce a rule selection algorithm called ROCCER, which operates by selecting classification rules from a larger set of rules – for instance found by Apriori – using ROC analysis. Experimental comparison with rule induction algorithms shows that ROCCER tends to produce considerably smaller rule sets with compatible Area Under the ROC Curve (AUC) values. The individual rules that compose the rule set also have higher support and stronger association indexes.

Ronaldo C. Prati, and Peter A. Flach. 2005. ROCCER: An Algorithm for Rule Learning Based on ROC Analysis. In Proceedings of the Nineteenth International Joint Conference on Artificial Intelligence, IJCAI-05. IJCAI, 823–827. https://ijcai.org/proceedings/2005/55

@inproceedings{ijcai2005p55,
  title     = {ROCCER: An Algorithm for Rule Learning Based on ROC Analysis},
  author    = {Prati, Ronaldo C. and Flach, Peter A.},
  booktitle = {Proceedings of the Nineteenth International Joint Conference on
               Artificial Intelligence, {IJCAI-05}},
  publisher = {International Joint Conferences on Artificial Intelligence},
  pages     = {823--827},
  year      = {2005},
  url       = {https://ijcai.org/proceedings/2005/55},
}

TY  - CPAPER
TI  - ROCCER: An Algorithm for Rule Learning Based on ROC Analysis
AU  - Prati, Ronaldo C.
AU  - Flach, Peter A.
T2  - Proceedings of the Nineteenth International Joint Conference on Artificial Intelligence, IJCAI-05
PB  - International Joint Conferences on Artificial Intelligence
PY  - 2005
SP  - 823
EP  - 827
UR  - https://ijcai.org/proceedings/2005/55
ER  -