Forecasting Pesticide Use on Golf Courses by Integration of Deep Learning and Decision Tree Techniques
Bibliographic record
Abstract
In the current study, a new hybrid machine learning (ML)-based model was developed by integrating a convolution neural network (CNN) with a random forest (RF) to forecast pesticide use on golf courses in Québec, Canada. Three main groups of independent variables were used to estimate pesticide use on golf courses, expressed as actual active ingredient rate (AAIR): (i) coordinates (i.e., longitude and latitude of the golf course), (ii) characteristics of the golf courses (i.e., pesticide type and the number of holes), and (iii) meteorological variables (i.e., total precipitation, P, and average temperature, T). The meteorological variables were collected from the Google Earth Engine by developing a JavaScript-based Code. On the basis of the different periods of total precipitation and average temperature, four different scenarios were defined. A data bank with more than 40,000 samples was used to calibrate and validate the developed model such that 70% of all samples were randomly selected to calibrate the model, while the remainder of the samples (i.e., 30%) that did not have any role in calibration were employed to validate the model’s generalizability. A comparison of different scenarios indicated that the model that considered the longitude and latitude of the golf course, pesticide type, and the number of holes in golf courses as well as total precipitation and average temperature from May to November as inputs (R = 0.997; NSE = 0.997; RMSE = 0.046; MAE = 0.026; NRMSE = 0.454; and PBIAS (%) = −0.443) outperformed the other models. Moreover, the sensitivity analysis result indicated that the total precipitation was the most critical variable in AAIR forecasting, while the average temperature, pesticide types, and the number of holes were ranked second to fourth, respectively.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".