Tuning Larger Membership Grades for Fuzzy Association Rules

Stephen G. Matthews

Research output: Chapter in Book/Report/Conference proceedingConference Contribution (Conference Proceeding)

58 Downloads (Pure)

Abstract

Sigma count measures scalar cardinality of fuzzy sets. A problem with sigma count is that values of scalar cardinality are calculated entirely from many small membership grades or entirely from few large membership grades. Two novel scalar cardinality measures are proposed for the fitness of a genetic algorithm for tuning membership functions prior to fuzzy association rule mining so that individual membership grades are larger. Preliminary results show a decrease in small membership grades and an increase in large membership grades for fuzzy association rules tested on real-world benchmark datasets.
Original languageEnglish
Title of host publicationIEEE International Conference on Fuzzy Systems (FUZZ-IEEE)
Pages1960
Number of pages1967
DOIs
Publication statusPublished - 2014

Fingerprint Dive into the research topics of 'Tuning Larger Membership Grades for Fuzzy Association Rules'. Together they form a unique fingerprint.

  • Cite this

    Matthews, S. G. (2014). Tuning Larger Membership Grades for Fuzzy Association Rules. In IEEE International Conference on Fuzzy Systems (FUZZ-IEEE) (pp. 1960) https://doi.org/10.1109/FUZZ-IEEE.2014.6891765