Automated H yperparameter Tuned Stacked Autoencoder based Rice Crop Yield Prediction Model

Rice crop production had greatly contributed to the food security of India, with over 40% of the entire crop production. Its productivity depends on favourable weather conditions. Changeability from season to season is unfavourable to the agronomist's livelihood and income. Enhancing the capability of agriculturalists for predicting crop production in distinct weather situations, can help farmers and other shareholders in making significant decisions concerning crop choice and agronomy. This estimation can assist agriculturalists and other shareholders to make better decisions on crop yielding. Prediction of crop yields using contemporary methods like Machine Learning (ML) techniques is in high demand for increasing the productivity of crops under several environmental circumstances. Therefore, this article develops an intelligent Hyperparameter Tuned Stacked Autoencoder Based Rice Crop Yield Prediction (HTSAE-RCYP) model. The proposed HTSAE-RCYP technique performs three stages of initial pre-processing: data conversion, null value removal, and data normalization. For the yield prediction process, the presented HTSAE-RCYP technique employs the SAE model which accurately forecasts the level of rice yield. To improve the predictive outcomes of the SAE model, a root means square propagation (RMS Prop) optimizer is utilized in this work. The experimental result analysis of the HTSAE-RCYP technique is tested using a series of simulations and the outcomes inferred the betterment of the HTSAE-RCYP technique over other existing models.