Fuzzy CoCo: a cooperative-coevolutionary approach to fuzzy modeling

Abstract

Coevolutionary algorithms have received increased attention in the past few years within the domain of evolutionary computation. In this paper, we combine the search power of coevolutionary computation with the expressive power of fuzzy systems, introducing a novel algorithm, Fuzzy CoCo: Fuzzy Cooperative Coevolution. We demonstrate the efficacy of Fuzzy CoCo by applying it to a hard, real-world problem-breast cancer diagnosis-obtaining the best results to date while expending less computational effort than formerly. Analyzing our results, we derive guidelines for setting the algorithm's parameters given a (hard) problem to solve. We hope Fuzzy CoCo proves to be a powerful tool in the fuzzy modeler's toolkit.

Designing a New Framework Using Type-2 FLS and Cooperative-Competitive Genetic Algorithms for Road Detection from IKONOS Satellite Imagery

“…This can be understood in two opposite ways: the individuals could collaborate for the same purpose and thus construct the solution together, or they could compete against each other for the same resources. The use of CCL algorithms is recommended when the following issues arise: the search space is huge, the problem may be decomposed into subcomponents or different coding schemes are used [62].…”

A coevolutionary algorithm for rules discovery in data mining

“…These result in the single population algorithm getting stuck in local optima, increase in computational complexity and thus poor explorative performance compared to its counterpart, i.e. coevolution algorithms that are able to handle these issues (Pen˜a-Reyes and Sipper 2001). When different types of fields exist, coevolution can be applied to encode the different types of fields with each field corresponding to one species.…”

“…One approach to ensure cohesiveness of the solutions is to evolve both elements simultaneously through coevolutionary-based algorithms, which could evolve multiple populations concurrently in data classification (Mendes et al 2001, Pen˜a-Reyes andSipper 2001). It has been shown that by coevolving a population of fuzzy membership function with a population of GA individuals (Pen˜a-Reyes and Sipper 2001) or GP tree individuals (Mendes et al 2001), better results could be produced compared to those without the coevolution.…”

Cooperative Coevolution for Learning Fuzzy Rule-Based Systems

“…Therefore, the use of cooperative coevolutionary algorithms is recommendable when the following issues arise [18]:…”

“…Actually, a method has been already proposed by Peña-Reyes and Sipper with this cooperative coevolutionary philosophy [18]. However, opposite to it, our proposal performs a more sophisticated learning of the RB based on the Cooperative Rules (COR) methodology [2], whose good performance is related to the consideration of cooperation among rules.…”