DefinePK

Abstract

Rainfall is one of the main attributes of climate changes in atmosphere. In this paper a method is proposed using Neural Network Backpropagation (BPP) for quantitative prediction of Rainfall rate. The architecture of Neural Network Backpropagation is built on N different attributes as input layer. This model is trained using five parameters as inputs in input layer and data over five years (2011-2016) received from Indian Meteorological Department. The configured Neural Network is applied on some portion of collected data of the state of Uttar Pradesh in India. In this study, we predicted the rainfall rate using Neural Network Backpropagation. The error is determined which can be less than the existing models and this achieved by handling outliers by applying K-Means algorithm which enhances the performance of Neural Network Backpropagation.