Crop Yield Prediction Using Hybrid Machine Learning Approach: A Case Study of Lentil (Lens culinaris Medik.)

This paper introduces a novel hybrid approach, combining machine learning algorithms with feature selection, for efficient modelling and forecasting of complex phenomenon governed by multifactorial nonlinear behaviours, such as crop yield. We have attempted to harness the benefits soft computing algorithm multivariate adaptive regression spline (MARS) selection coupled support vector (SVR) artificial neural network (ANN) efficiently mapping relationship between predictors predictand variables using MARS-ANN MARS-SVR frameworks. The performances are com-pared on different fit statistics RMSE, MAD, MAPE, etc., numeric agronomic traits 518 lentil genotypes predict grain proposed MARS-based models outperformed individual MARS, SVR ANN. is largely due enhanced ex-traction capability MARS model ability ANN SVR. superiority MARS-SVM in terms building generalisation was demonstrated.