Evolutionary Support Vector Machines for Diabetes Mellitus Diagnosis

The aim of this paper is to validate the new paradigm of evolutionary support vector machines (ESVMs) for binary classification also through an application to a real-world problem, i.e. the diagnosis of diabetes mellitus. ESVMs were developed through hybridization between the strong learning paradigm of support vector machines (SVMs) and the optimization power of evolutionary computation. Hybridization is achieved at the level of solving the constrained optimization problem within the SVMs, which is a difficult task to perform in its standard manner. ESVMs have been so far applied to the binary classification of two-dimensional points. In this paper, experiments are conducted on the benchmark problem concerning diabetes of the UCI repository of machine learning data sets. Obtained results proved the correctness and promise of the new hybridized learning technique and demonstrated its ability to solve any case of binary standard classification