Ensemble neural network optimization using a gravitational search algorithm with interval type-1 and type-2 fuzzy parameter adaptation in pattern recognition applications

Abstract

In this paper we consider the problem of optimizing ensemble neural networks for pattern recognition with Type-1 and Type-2 fuzzy logic for parameter adaptation in the gravitational search algorithm. The database to be used is of echocardiography images, since these images are very important in clinical echocardiography, and these images help the doctors to diagnose cardiac diseases, as well as to prevent this type of diseases in patient treatment.