Henry Gas Solubility Optimization with Deep Learning Model for Crop Yield Prediction and Crop Type Classification

Agriculture is considered to be the backbone of the Indian economy, with more than half of the country?s population depending on agriculture. Crop production can be forecasted by utilizing machine learning (ML) methods rely upon parameters like meteorological conditions, rainfall, and crop. The powerful and most popular supervised ML technique, Random Forest, can do both regression and classification tasks. It can be used in crop selection for reducing crop yield output losses, irrespective of the distracting atmosphere. Meteorological conditions and other related environmental components bring a significant danger to the long-term viability of agriculture. ML is significant as renders a decision-support tool for Crop Yield Prediction (CYP), which will help to make decisions like which crops have to be cultivated and during the crop's growing season. This manuscript develops a new Henry Gas Solubility Optimization with Deep Learning Model (HGSO-DLM) technique to predict crop yield and classify crop types. In the presented HGSO-DLM model, two major processes are involved. At the initial level, the presented HGSO-DLM model employs a deep stacking auto-encoding (DSAE) model for yield prediction and crop classification. Next, in the second stage, the HGSO algorithm is applied for effectual hyper parameter optimization of the DSAE model. To exhibit the improvements of the HGSO-DLM model, a wide range of simulation results were performed and the comparison study reported the improvements of the HGSO-DLM model.