Comparative study of neural network architectures for rainfall prediction

Majority of Indian farmers depend on rainfall for agriculture. Thus, in an agrarian country like India, rainfall prediction becomes very important. This paper presents comparative study of neural network architectures namely Back Propagation Neural Network (BPNN), Generalized Regression Neural Network (GRNN) and Radial Basis function Neural Network (RBNN) to predict rainfall in Thanjavur district of southern province TamilNadu, India. The different models are trained using the training data set and have been tested for accuracy on available test data. MATLAB has been used for model development. After training all networks and testing them we found that RBNN gives best result for prediction.