Prediction of Crop Cultivation

Farmers have certain expectations of how much crop will get and make financial decisions based on it. In the past, farmers used to predict crops based on their own experience and observed weather conditions. Weather, pests, and harvest operation may be kept as reference for future years. Currently software can augment traditional knowledge. Keeping accurate data is an important aspect of agricultural risk management. We propose to use machine-learning techniques to develop a prediction model for crop yield production. We compare the performance of various linear and non-linear regressor models using 5-fold cross validation. We found that using mostly default settings, the random forest regressor performed the best, followed by nearest-neighbor regression, L
²
linear regression with polynomial features, and support-vector regression using a Radial Basis Function (RBF) kernel.