Application of NARX Neural Network for Predicting Suitable crop for Cultivation- An Comparative analysis

Crop selection is an essential factor in crop farming success. It is a requirement that must be fulfilled before launching an agricultural business. Crop yield depends on various complex parameters such as genotype, climatic factors, soil nutiients and interaction with other environmental parameters. A fundamental understanding of the functional relationship between yield and these interaction components is required for accurate yield and crop selection. This calls for both huge datasets and powerful algorithms to uncover. Machine learningis a powerful tool for forecasting crop yields, and determining what crops to sow and what to do during the crop’s growing season. A variety of machine learning approaches have been applied to improve crop yield prediction research. In this case, predictive maintenance techniques with Artificial Neural Networks (ANNs) have assisted decisionmakers in selecting the best crop. For crop yield forecasting regarding to crop selection criteria, this study offers a Nonlinear Autoregressive with Exogenous Input(NARpANN. The performance measure Mean Squared Error (MSE) and correlation coefficient (R) were determined to assess the NARX models that had been built by choosing the proper crop for the specific soil conditions and climate, one can increase yields while reducing irrigation waterusage.