Indian Journal of Science and Technology

Abstract

Data mining is an emerging technology for applications such as text based mining, web based mining and it performs a major role in various domains for numerical data analysis, data statistics and machine learning. In this paper, data mining is used in machine learning of ANN (Artificial Neural Network). The Pruning technique in an MLP (The Multilayer Perceptron) neural network is to remove the unwanted neurons based upon their corresponding weights; as a result it improves the accuracy and speed of the network. In the existing system, based on their synaptic weights the pruning is performed and it removes the lowest weight neuron form the network. The result obtained from the existing method does not produce an optimized removal of the neuron. In the proposed system pruning is performed by using divisive clustering in MLP neural network. The main purpose of the Divisive algorithm in ANN is to split each neuron weight into sub neuron up to the fixed level and then remove the least weighted hidden neuron. The proposed method is implemented using the Java language. The Performance result obtained from the proposed method shows that it reduces the error rate and improves efficiency and accuracy of the MLP network. The present results confirm that DI-ANN (Divisive Artificial Neural Network) can provide a fast, accurate, and consistent methodology applicable to the neural network
Keywords: DI-ANN (Divisive Artificial Neural Network) Algorithm, MLP (Multi Layer Perceptron), Neural Network (NN), Pruning Method