Techniques for Developing Large Scale Fuzzy Logic Systems

Jon C. Ervin, Sema E. Alptekin.

In this paper, we describe various techniques used to make Intuitionistic Fuzzy Logic Systems amenable to operating on applications with large numbers of inputs. A rule reduction technique known as Combs method is combined with an automated tuning process based on Particle Swarm Optimization. A second stage of tuning on rule weights results in improved performance and further reduction in the size of the rule-base. The entire process has been developed to operate within the Matlab software environment. The technique is tested against the Wisconsin Breast Cancer Database. The use of these tools shows great promise in significantly expanding the range and complexity of problems that can be addressed using Atanassov’s Intuitionistic Fuzzy Logic.

PDF full paper