Evolving Artificial Neural Networks Using Opposition Based Particle Swarm Optimization Neural Network for Data Classification

Artificial neural network (ANN) has a wide variety of practice for the solution of problems in the area of data classification. Back propagation algorithm is a famous neural network (NN) traditional training approach. Since this classical training technique has many drawbacks like stuck in the local minima, maximum number of iterations required, in this paper the training of the NN has been implemented with the opposition based with particle swarm optimization neural network (OPSONN) algorithm. These algorithms that are used for the NN training can be applied for the solutions of data classification problems. It is renowned that different techniques comparison is also as vital as by proposing a new technique for data classification. In this paper, a detailed comparative performance analysis for the training of neural network is observed on the different data sets taken from UCI repository. Results demonstrates that opposition based particle swarm optimization neural network (OPSONN) may offer efficient and best substitute to traditional training approach of the neural network for the solution of problems of data classification. The results are compared with OPSONN learning algorithm for feed forward neural network (FNN). The subsequent exactness of FNNs trained with PSO (PSONN), back propagation algorithm (BPA), and back propagation algorithm with momentum is likewise examined. The trial results demonstrate that OPSONN outperforms PSONN, back propagation algorithm (BPA), and back propagation algorithm with momentum for preparing FFNNs as far as accuracy rate and better precision. It is likewise demonstrated that an FFNN prepared with OPSONN technique has preferable exactness over one trained with different methods.