Supplemental material for: Detecting intramammary infection at dry-off integrating sensor data and machine learning algorithms
Description
Supplemental Tables S1 to S3 for a study on the detection of intramammary infection at dry-off in dairy cows using routinely available farm data (Dairy Herd Improvement records, milking-parlor sensors, collar sensors, genetic evaluations and temperature-humidity index) and 6 machine learning algorithms. Data were collected on a commercial Holstein Friesian farm in the Veneto region of Italy (294 cows, November 2023 to August 2024). Table S1: descriptive statistics (mean, SD, median, missing %) of the continuous candidate predictors by cow-level infection status. Table S2: distribution of the categorical candidate predictors by cow-level infection status. Table S3: calibration intercept, calibration slope, Brier score and mean predicted probability of the 6 models on the held-out test set (90 cows). Variable script names correspond to those used in the manuscript.
Files
Institutions
- University of PaduaVeneto, Padua