MLP for prediction of area and rice production of upper Brahmaputra Valley zone of Assam

Abstract

Listen

Translate article

The present study is carried out to predict the area and Rice production of Upper Brahmaputra Valley Zone of Assam using Artificial Neural Network (ANN). Multilayer Perceptron (MLP) with single hidden layer has been trained with the secondary data of the area, Rice production and meteorological data. Area and Rice production data are collected from the Directorate of Economics and Statistics, Directorate of Agriculture, Government of Assam, Guwahati and Meteorological data are obtained from National Data Centre, India Meteorological Department (IMD), Pune; Regional Meteorological Centre, IMD, Guwahati and Department of Agrometeorology, Assam Agricultural University, Jorhat, Assam. The appropriate model for each of the network is identified. The performance of the developed ANN model viz. MLP with Backpropagation Algorithm has been measured using Root Mean Squared Errors (RMSE) and Correlation Coefficients (CC). The accuracy of the developed MLP model has been compared with Multiple Linear Regression (MLR) Model and experimental results show MLP model outperforms MLR model. Sensitivity analysis has been performed for Prediction of Summer Rice production and results show that technology index is the most sensitive parameter for Summer Rice production followed by rainfall index for Upper Brahmaputra Valley Zone of Assam.