Boosting a genetic fuzzy classifier

Author

Hoffmann, Frank

Author_Institution

R. Inst. of Technol., Stockholm, Sweden

Volume

3

fYear

2001

fDate

25-28 July 2001

Firstpage

1564

Abstract

The paper presents a novel boosting algorithm for genetic learning of fuzzy classification rules. The method is based on the iterative rule learning approach to fuzzy rule base system design. The fuzzy rule base is built in an incremental fashion, in that the evolutionary algorithm extracts one fuzzy classifier rule at a time. The boosting mechanism reduces the weight of those training instances that are classified correctly by the new rule, such that the next iteration of the evolutionary algorithm focuses the search on those fuzzy rules that capture the currently uncovered or misclassified instances. The weight of a fuzzy rule reflects the relative strength the boosting algorithm assigns to the rule class when it aggregates the casted votes. The method is applied to the Wisconsin breast cancer diagnosis data set

Keywords

fuzzy logic; fuzzy set theory; genetic algorithms; knowledge based systems; learning (artificial intelligence); search problems; Wisconsin breast cancer diagnosis data set; boosting algorithm; boosting mechanism; casted votes; evolutionary algorithm; fuzzy classification rules; fuzzy classifier rule extraction; fuzzy rule base system design; genetic fuzzy classifier boosting; genetic learning; incremental design; iterative rule learning approach; rule class; search; training instances; Aggregates; Boosting; Evolutionary computation; Fuzzy sets; Fuzzy systems; Genetics; Iterative algorithms; Iterative methods; Training data; Voting;

fLanguage

English

Publisher

ieee

Conference_Titel

IFSA World Congress and 20th NAFIPS International Conference, 2001. Joint 9th

Conference_Location

Vancouver, BC

Print_ISBN

0-7803-7078-3

Type

conf

DOI

10.1109/NAFIPS.2001.943782

Filename

943782