Intelligent Crop Recommendation with Yield Prediction using Dragonfly Algorithm based Deep Learning Model

Most developing countries like India considered agriculture a primary source of food, income, livelihood, and employment. The absence of adequate technical knowledge and irregular weather conditions resulted in reduced annual crop productivity. Hence, the selection of proper crops for improved productivity and yield becomes essential to fulfilling the rising food demand globally. In addition, the forecasting of crop yield is also necessary and it depends upon various parameters such as type of irrigation, weather, region, etc. The latest developments in Machine Learning (ML) and Deep Learning (DL) techniques allow for designing effective crop recommendation and yield prediction models. With this motivation, this article concentrates on the design of intelligent crop recommendation and yield prediction using dragonfly algorithm based deep learning (ICRYP-DFADL) technique. The presented ICRYP-DFADL model majorly aims to choose proper crops and forecast the crop yield proficiently. To do so, the ICRYP-DFADL model primarily utilizes DFA with deep neural network (DNN) model for crop recommendation. The design of DFA helps to appropriately select the DNN hyperparameters. In addition, the cascaded recurrent neural network (CRNN) model is employed for effectual prediction of crop yield. Finally, the experimental validation of the ICRYP-DFADL model is tested using distinct dataset. The experimental values pointed out that the ICRYP-DFADL model has shown promising results over other models.