An Empirical Study on Neuroevolutional Algorithm Based on Machine Learning for Crop Yield Prediction

Abstract

Machine learning has been come out with high performance computation power leads to create a great prospect in multi-disciplinary domain. Here, we present a novel machine learning method for predicting crop yield. The classification technique using machine learning algorithm demonstrated the performance improvement in prediction of crop yield. It depends on the factors of weather which have relationship with climate change data, soil of that area, and water irrigations. Here, we have illustrated an approach of implementing neuroevolution model based on ANN for predicting wheat crop yield. Crop yield prediction at different months is considered from June to September; the yields predictions are computed based on weather and fertilizer utilized data. A major improvement in the prediction ability is observed that yield diverge as for the season changes based on weather data. Therefore, the result of the proposed model assists in decision making in advance for planting wheat crop. The outcomes are more functional for decision making as well as in transplantation of wheat in advance with various farm activities throughout various stages of the wheat crop growing. Also, the same model can also be utilized for predicting various agricultural data such as disease prediction and weather prediction.