Forecasting of powdery mildew in mustard (Brassica juncea) crop using artificial neural networks approach

Abstract

Recently artificial neural networks (ANNs) techniques has become the focus of much attention, largely because of their wide range of applicability and the ease with which they can treat complicated problems even if the data are imprecise and noisy. From statistical perspective, neural networks are interesting because of their potential use in prediction. This methodology have been illustrated by considering various aspects, viz maximum pest disease severity, crop age at first appearance of disease, crop age at maximum disease severity for powdery mildew in mustard crop at S K Nagar (Gujarat) as response variable and weather indices (a technique based on relatively smaller number of manageable variables and at the same time taking care of entire weather distribution) as predictors. In this study, data have been taken from Mission mode project under National Agricultural Technology Project, entitled ‘Development of weather-based forewarning system for crop pests and diseases’, CRIDA, Hyderabad in which the IASRI was one of the cooperating institutions. Two type of neural network architecture namely Multilayer perceptron (MLP) and Radial basis function (RBF) were attempted and compared with weather indices based regression model and it has been found that a MLP performs best in terms of mean absolute percentage error (MAPE).