An analysis of particle swarm optimization for feature selection on medical data

Classification of medical data to determine different type or subtype of medical condition is significantly important to research disease in level. Feature selection technologies used to reduce features numbers and find informative features have been presented in recent years. But the performance of feature selection in medical data classification research is still legendary. In this study, different classification algorithms are established to classify the different medical datasets and compared the results obtained by using the particle swarm optimization (PSO), the classification accuracy. Particle swarm optimization is a developed algorithm that explains the movement of group of birds in space in mathematical terms. In PSO, each potential problem is viewed as an element with specific velocity that flies through a problem space as if it was a flock of bird. In this paper we present a literary review of papers on PSO and record its way through inception and implementing it in different physical problems. We present a comparative table of implementation for PSO and review PSO success in various fields of science. The suggested method is tested on several standard data sets at UCI database and its performance is compared with those of Particle Swarm Optimization (PSO) using classifiers. This paper motivates readers to join the PSO world.