Novel Hybrid Statistical Learning Framework Coupled with Random Forest and Grasshopper Optimization Algorithm to Forecast Pesticide Use on Golf Courses

Golf course maintenance requires the use of several inputs, such as pesticides and fertilizers, that can be harmful to human health or environment. Understanding factors associated with pesticide on golf courses may help golf-course managers reduce their reliance these products. In this study, we used a database about 14,000 applications in province Québec, Canada, develop novel hybrid machine learning approach predict courses. We created proposed model, called RF-SVM-GOA, by coupling support vector (SVM) random forest (RF) grasshopper optimization algorithm (GOA). applied RF handle wide range datasets GOA find optimal SVM settings. considered five different dependent variables—region, ID, number holes, year, treated area—as input variables. The experimental results confirmed developed RF-SVM-GOA was able estimate active ingredient total (AIT) high level accuracy (R = 0.99; MAE 0.84; RMSE NRMSE 0.04). compared produced model those four tree-based techniques including M5P, tree, reduced error pruning tree (REP tree), RF, well two non-tree-based generalized structure group method data handling (GSGMDH) evolutionary polynomial regression (EPR). computational showed higher, outperforming other methods. analyzed sensitivity most effective variables AIT forecasting. indicated area is variable current study provide for increasing sustainability management.