Automating classification of veterinary biosecurity recommendations using machine learning
Bibliographic record
Abstract
ProAction® is a mandatory Canadian milk quality assurance program that requires dairy farmers to conduct a biosecurity risk assessment with a veterinarian. Based on this assessment, the veterinarian provides personalized recommendations to improve farm biosecurity, resulting in a large volume of text data. This study aimed to develop a machine learning model capable of automatically classifying these biosecurity recommendations into 12 predefined categories. As the recommendations were written in French or English, all texts were translated into French to ensure consistency in feature extraction and model training. The model was trained on 11,250 manually classified veterinary recommendations from 3825 Québec dairy herds, collected between 2018 and 2021. Three algorithms were tested: Multinomial Naïve Bayes (MNB), Support Vector Machine (SVM), and Random Forest (RF). Performance was evaluated using precision, recall, and F1-score. The SVM achieved the highest performance while maintaining efficient processing time. The trained SVM model was selected to classify new recommendations collected between 2022 and 2024 from herds in Alberta, Ontario and Québec. To evaluate model's performance on this new dataset, a random sample of 250 recommendations was manually classified. The agreement between human classification and model predictions resulted in a Cohen's Kappa of 0.88, suggesting strong agreement. This study highlights the potential of machine learning to classify biosecurity recommendations and support timely, informed decision-making in dairy herd management.
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 | 0.001 | 0.005 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.002 |
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".