Fuzzy Rule Extraction

The goal is to define innovative techniques and approaches for extracting novel knowledge, expressed in the form of fuzzy rules, from data samples when no assessed knowledge is available or in order to complete that existing one. Our approach relies on a six-steps data driven methodology to automatically build fuzzy DSSs, where each step can be approached using several strategies. The approach is based on the extraction of crisp rules from data, their fuzzification and the successive optimization with respect to the membership functions of the linguistic variables involved in the rules. In this approach, the number of rules selected is user-defined and the optimal number of rules for the tested dataset is determined by exploiting not only the correct classification rate but also a confidence based criterion aimed at obtaining highly understandable systems.

Publications

  • A. d'Acierno, G. De Pietro, M. Esposito, "Data Driven Generation Of Fuzzy Systems: An Application To Breast Cancer Detection", in the Seventh International Meeting On Computational Intelligence Methods For Bioinformatics And Biostatistics (CIBB 2010), Palermo, Italy, September 16-18, 2010.
  • M. Esposito, G. De Pietro, A. D' Acierno, "Eliciting fuzzy knowledge from the PIMA dataset", Joint NETTAB 2010 and BBCC 2010 workshops focused on Biological Wikis, Napoli, Italia, 2010.
  • A. d'Acierno, G. De Pietro, M. Esposito, Introducing jFKD: a java Fuzzy Knowledge Discoverer In NETTAB 2011 workshop focused on Clinical Bioinformatics October 12-14, 2011, Pavia, Italy.