A new harmony search algorithm with evolving spiking neural network for classification problems

Abstract

In this study, a new hybrid harmony search algorithm with evolving spiking neural network (NHS-ESNN) for classification issues has been demonstrated. Harmony search has been used to enhance the standard ESNN model. This new algorithm plays an effective role in improving the flexibility of the ESNN algorithm in creating superior solutions to conquer the disadvantages of ESNN in determining the best number of pre-synaptic neurons which is necessary in constructing the ESNN structure. Various standard data sets from UCI machine learning are utilised for examining the new model performance. It has been detected that the NHS-ESNN give competitive results in classification accuracy and other performance measures compared to the standard ESNN. More argumentation is provided to verify the effectiveness of the new model in classification issues.