Comparative analysis of backpropagation and RBF neural network on monthly rainfall prediction

Rainfall plays a major role in the economic growth of every country. It is an essential component in the human life cycle. Its accurate and predetermined forecast can prevent many natural hazards and helpful to the tourists for travelling beautiful destinations of the world. Due to the current climatic variations, the accuracy of forecasting rainfall is very important. In this paper, we have used back propagation and radial basis function neural network techniques for predicting the monthly rainfall values based on the data collected from the Coonoor region in the Nilgiri district (Tamil Nadu). Furthermore, mean square error and accuracy are the performance indices used for the comparative analysis. Experimental results showed that the radial basis function neural network acts as a better prediction model with smaller mean square error and higher accuracy values. Furthermore, we have used these techniques for predicting the future data based on the previously collected data.