Machine learning algorithms for clinical mastitis prediction in a dairy herd using automated milking system data

Faris DN, Gad AM, El-Tarabany MS, Ramadan SI, Afifi GG, Manaa EA. Machine learning algorithms for clinical mastitis prediction in a dairy herd using automated milking system data. J Adv Vet Res. 2024;14(6).

This study compares six machine learning algorithms for predicting clinical mastitis in dairy cows based on data from an automated milking system. The study involved 1493 cows with clinical mastitis and 2387 healthy cows, with performance metrics such as accuracy, precision, recall, F1-score, and area under the curve (AUC) used to evaluate the models. The Decision Tree algorithm and Gaussian Naïve Bayes scored the highest AUC of 71%, with the Decision Tree showing greater stability and accuracy (73%). Key features influencing the Decision Tree model included days in milk, age, lactation order, mature herd equivalent, and average daily milk yield.