Authors : Neeta Verma, Y.D.S Arya and K.C. Tripathi
Abstract: There is great regional and temporal variation in the distribution of rainfall. The great variation in the amount of rainfall both spatially and temporally, the high degree of uncertainty related to the date of arrival, etc., are unexplained. A neural network model for analyzing the all India rainfall has been developed by using the 142 years of data. In formulating artificial neural network based predictive models three layered network has been constructed. The models under study are different in the number of hidden neurons. The main objective of this study is to evaluate the applicability of ANN. The performance of different networks have been evaluated and tested. The reason of using the ANN (Artificial Neural Network) model is based on prediction by smartly analyzing the trend from the previously existing data set. In the present research, the last 142 years data of all Indian rainfall has been analyzed through artificial neural network models. The Artificial Neural Network (ANN) technique with back-propagation algorithm for the predictability of AIR with 1 lag by analysing the historical time series of 142 years of AIR data. The ANN model used to forecast rainfall that is validated using the correlation coefficient and Root Mean Square Error (RMSE). The statistical parameters are not sufficient for model accuracy, so for accuracy skill scores for verification areobtained in this study.
Neeta Verma, Y.D.S Arya and K.C. Tripathi, 2017. Skill Scores Verification for all India Rainfall Data Using Artificial Neural Network. International Journal of Soft Computing, 12: 13-19.