Growth Indexes and Yield Prediction of Summer Maize in China Based on Supervised Machine Learning Method

Leaf area index and dry matter mass are important indicators for crop growth yields. In order to solve the problem of predicting summer maize yield under different soil quality field management conditions, this study proposes a prediction model based on supervised machine learning regression algorithm. Firstly, data pool was constructed by collecting measured in main planting area. The total water input (rainfall plus irrigation water), fertilization, quality, density were selected as training set. Then, maximum leaf (LAImax), material (Dmax), yields (Y) trained using Gaussian (rational quadratic kernel function Matern function), support vector (SVM) linear models. models verified with data-set not included pool, fertilizer coupling functions developed. results showed that compared models, comprising rational had good accuracy. coefficients determination (R2) 0.91, 0.89 0.88; root-mean-square errors (RMSEs) 0.3, 1138.6 666.16 kg/hm2; relative (rRMSEs) 6.3%, 5.94% 6.53% LAImax, Dmax Y, respectively. optimal inputs nitrogen applications indicated consistent range from tests. algorithm provides simple method predict optimize only density.