Genetic Algorithm and Firefly Algorithm in a Hybrid Approach for Breast Cancer Diagnosis

Feed-forward neural networks are popular classification tools which are broadly used for early detection and diagnosis of breast cancer. In recent years, a great attention has been paid to bio-inspired optimization techniques due to its robustness, simplicity and efficiency in solving complex optimization problems. In this paper, it is intended to introduce a Genetic Algorithm based Firefly Algorithm for training neural networks. The proposed algorithm is used to optimize the weights between layers and biases of the neuron network in order to minimize the fitness function which is defined as the mean squared error. The simulation results indicate that better performance of the Firefly Algorithm in optimizing weights and biases is obtained when being hybridized with Genetic Algorithm. The proposed algorithm has been tested on Wisconsin Breast Cancer Dataset in order to evaluate its performance and the efficiency and effectiveness of the proposed algorithm by comparing its results with the existing methods. The results of the proposed algorithm were compared with that of the other techniques Firefly Algorithm, Biogeography Based Optimization, Particle Swarm Optimization and Ant Colony Optimization. It was found that the proposed Genetic Algorithm based Firefly Algorithm approach was capable of achieving the lowest mean squared error of 0.0014 compared to other algorithms as mean squared error values for other algorithms were 0.002 for Firefly Algorithm, 0.003 for Biogeography Based Optimization, 0.0135 for Ant Colony Optimization , 0.035 for Particle Swarm Optimization.