Enhanced Crop Yield Prediction using Hybrid Deep Learning Models

India's economy depends primarily on agriculture. Crop yield offers significant potential for food production worldwide. It enables farmers and policymakers to make informed decisions to improve food security and maximize resource allocation. The yield of crops depends heavily on multiple factors such as climate conditions, rainfall, soil nutrients, agricultural practices, etc. The climate has a major impact on the crop's potential production. A hybrid LSTM model has been developed for crop yield prediction and trained on a diverse dataset comprising historical yield records and meteorological data. Experimental results show that hybrid LSTM models outperform the standalone models, such as CNN, GRU, and LSTM, in crop yield prediction, and the model performance is measured using MSE, RMSE, MAE, and R2. The Hybrid LSTM Model has an MSE of 0.0015, RMSE of 0.03, MAE of 0.007, and Rsquared value of 0.86.