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

Abstract

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

Forecasting Pesticide Use on Golf Courses by Integration of Deep Learning and Decision Tree Techniques

“…(1) precise estimation of the AAIR, (2) automatic classification of the data to overcome existing problems in the range of the AAIR values (the third quartile of the AAIR was lower than the average of all samples, see Figure 3), and (3) testing of different scenarios combining golf course characteristics and meteorological variables, as was recommended in a recently published paper [9]. The developed model based on the selected scenario (i.e., S1) can be used as a practical tool for forecasting AAIR for future years on the basis of the different climate change scenarios.…”

“…Therefore, it was time-consuming to model many samples, especially for the current study for which modeling needed a high number of iterations to achieve the most optimum results in all automated created groups of samples by the RF. The main modeling challenges for a computer with an Intel Core i7 processor and 16 GB RAM were (1) the high number of iterations required for the developed hybrid model, (2) low memory in using the newly developed hybrid machine learning model for the forecasting of the golf course characteristics [9], and (3) finding the optimum values of the developed RFCNN.…”

“…In a previous study [9], we used the database of pesticide applications compiled by the MELCC from 2003 to 2018 to develop a machine learning algorithm that can reliably predict pesticide use on golf courses in Québec. This algorithm relied on a hybrid approach combining a tree-based method (random forest-RF) and a non-tree-based method (support vector machine-SVM) followed by an optimization phase (grasshopper optimization algorithm-GOA).…”

The importance of open data describing prey item species lists for endangered species

“…Fly, grasshopper and species of cricket are strongly influenced by the availability of herbaceous spring and summer vegetation and flowering plants (Ibanez et al, 2013; Reemer & Rotheray, 2009). These three insect prey items are also sensitive to pesticide practices in agricultural and human‐use lands (Grégoire et al, 2022), and thus, negative spillover from crop management needs to be considered as a key rationale for diet data inclusion into planning decisions (Montoya et al, 2021). These data suggest that planning for habitat conservation and restoration with the endangered species G. sila as an indicator for the region must incorporate direct needs such as shrubs (for shelter) but also ensure that other vegetation is present to support its prey items (for their diets).…”