Evolutionary Fuzzy ARTMAP Approach for Breast Cancer Diagnosis

The objective of this paper is to present the strength of fuzzy artmap which is kind of neural networks in the medical field by improving its performance by genetic algorithm. Fuzzy ARTMAP is both much faster and incrementally stable than the other ordinary neural networks models like Multilayer Perceptron. Fuzzy artmap’s parameters have legal range of values that should be determined in the simulation. These parameters should be adjusted and tuned many times to get the best results and optimum solution in order to generate accurate classification system. As there are large ranges of parameter’s values so that lead to huge number of possible solutions. The problem is to try to find the best solution among whole possibilities in the problem search space and each point in the search space represents one feasible solution. The proposed solution to this problem is using genetic algorithms to optimize fuzzy artmap parameters and this yield to improving fuzzy artmap performance. Genetic algorithms are a part of evolutionary computing which can be used to quickly scan a vast solution set. This enhanced approached evolved fuzzy artmap will be used to generate breast cancer diagnosis system.