Agriculture Crop Yield Prediction Using Deep Learning Models

Abstract

Crop yield prediction is a big challenge in agricultural research. Due to the natural calamities, the farmers were not able to predict their crop yields. Hence, the prediction methodology is necessary for the researchers to identify the productivity and demand of the particular crop. Innovation in crop yield prediction models and methods can assist researchers in finding better results. The various machine learning (ML) models have been developed, and their performance has been evaluated through different research with real-time agricultural datasets. But still, the performance of the ML models is not satisfactory, and hence an improvement is needed in some factors. In this research, deep learning-based algorithms such as Deep Neural Network (DNN), Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), and Long Short Term Memory (LSTM) were used to evaluate the performance of the model for crop yield prediction. In this research, various experiments were performed using the DNN, CNN, RNN, and LSTM models based on the agricultural dataset. The proposed models were compared with the various features, and the Long Short Term Memory (LSTM) algorithm gave the best accuracy among the other models.