Articles | Volume 69, issue 3
https://doi.org/10.5194/aab-69-469-2026
https://doi.org/10.5194/aab-69-469-2026
Original study
 | 
28 Aug 2026
Original study |  | 28 Aug 2026

Assessment of reproductive performance in dairy cows using explainable machine learning

Elif Çelik Gürbulak, Uğur Kara, Esra Canooğlu, Hazal Aysın Arslan, Ece Çetin, Mehmet Demirel, and Kutlay Gürbulak

Related authors

Comparative analysis of serum anti-Müllerian hormone (AMH) levels in sheep: the role of genetic background and physiological status
Uğur Kara and Davut Koca
Arch. Anim. Breed., 69, 37–44, https://doi.org/10.5194/aab-69-37-2026,https://doi.org/10.5194/aab-69-37-2026, 2026
Short summary

Cited articles

Alpan, O. and Arpacık, R.: Sığır yetiştiriciliği, 2nd edn., Şahin Matbaası, Türkiye, 1998. 
Angelov, P. P., Soares, E. A., Jiang, R., Arnold, N. I., and Atkinson, P. M.: Explainable artificial intelligence: an analytical review, Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 11, e1424, https://doi.org/10.1002/widm.1424?utm_source=chatgpt.com, 2021. 
Borș, S. I. and Borș, A.: Ovarian cysts, an anovulatory condition in dairy cattle, J. Vet. Med. Sci., 82, 1515–1522, https://doi.org/10.1292/jvms.20-0381, 2020. 
Bruinjé, T. C. and LeBlanc, S. J.: Inflammation and health in the transition period influence reproductive function in dairy cows, Animals, 15, 633, https://doi.org/10.3390/ani15050633, 2025. 
Burton, J., Farrell, S., Mäntylä Noble, P.-J., and Al Moubayed, N.: Explainable text-tabular models for predicting mortality risk in companion animals, Sci. Rep., 14, 14217, https://doi.org/10.1038/s41598-024-64551-1, 2024. 
Download
Short summary

This study demonstrates the applicability of an explainable machine learning framework (XGBoost + SHAP) for investigating reproductive performance in dairy cows. Although the predictive performance of the independent test dataset was limited (R² = 0.01), SHAP provided transparent global- and individual-level explanations of model predictions, highlighting the methodological potential of explainable artificial intelligence for veterinary data analysis.

 
 
Share