Artificial neural networks for predicting first-lactation 305-day milk yield in crossbred cattle

This study was conducted using the first-lactation records of 1092 Vrindavani crossbred cattle to compare the relative efficiency of an artificial neural network (ANN) versus multiple linear regression for predicting the first-lactation 305-day milk yield (FL305DMY). The two input sets used for pre...

Full description

Saved in:
Bibliographic Details
Main Authors: SM Usman, A Kumar, A Sherasiya, IM Youssef, MM Abo Ghanima, J Chandrakar, R Tiwari, NP Singh, QS Sahib, T Dutt, AA Swelum
Format: Article
Language:English
Published: South African Society for Animal Science 2025-07-01
Series:South African Journal of Animal Science
Subjects:
Online Access:https://www.sajas.co.za/article/view/23007
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849411381460205568
author SM Usman
A Kumar
A Sherasiya
IM Youssef
MM Abo Ghanima
J Chandrakar
R Tiwari
NP Singh
QS Sahib
T Dutt
AA Swelum
author_facet SM Usman
A Kumar
A Sherasiya
IM Youssef
MM Abo Ghanima
J Chandrakar
R Tiwari
NP Singh
QS Sahib
T Dutt
AA Swelum
author_sort SM Usman
collection DOAJ
description This study was conducted using the first-lactation records of 1092 Vrindavani crossbred cattle to compare the relative efficiency of an artificial neural network (ANN) versus multiple linear regression for predicting the first-lactation 305-day milk yield (FL305DMY). The two input sets used for predicting FL305DMY in the study were input set-1: first four monthly test-day milk yields, age at first calving, and peak milk yield; and input set-2: first four monthly milk yields, age at first calving, and peak milk yield. The ANN was trained using a backpropagation algorithm based on Bayesian regularisation, and the algorithm was tested using four sets of training and test data at ratios of 66.67:33.33, 75:25, 80:20, and 90:10. The results revealed that the coefficient of determination showed no regular trend with decreasing the test dataset. Nevertheless, the observed values were highest for the 90:10 ratio of training-test data for both input sets, with the lowest root mean square error. The ANN model outperformed the multiple linear regression model when predicting FL305DMY, with an accuracy of 79.09% for input set-1 and 83.67% for input set-2, with the lowest root mean square error values for both input sets. Therefore, the ANN model can be used as an alternative technique to predict FL305DMY in Vrindavani cows. Submitted 30 June 2024; Accepted 20 December 2024; Published January 2025 ------------------------------------------------------------------ Significance of research to South African science The research demonstrates the potential of artificial neural networks (ANNs) to enhance dairy herd management through accurate early prediction of first-lactation 305-day milk yield in crossbred cattle. The study compares ANN and multiple linear regression models, showing that ANNs provide superior predictive accuracy and lower error margins. This has important implications for local dairy farming systems, particularly in data-limited environments, as it offers a practical tool for early identification of high- and low-producing cows. By enabling data-driven selection and culling decisions, the approach supports improved productivity, genetic progress, and economic efficiency in the South African livestock sector.
format Article
id doaj-art-c1b38f435b764b0b89cf16f33f8cfde0
institution Kabale University
issn 0375-1589
2221-4062
language English
publishDate 2025-07-01
publisher South African Society for Animal Science
record_format Article
series South African Journal of Animal Science
spelling doaj-art-c1b38f435b764b0b89cf16f33f8cfde02025-08-20T03:34:47ZengSouth African Society for Animal ScienceSouth African Journal of Animal Science0375-15892221-40622025-07-0155110.4314/sajas.v55i1.01Artificial neural networks for predicting first-lactation 305-day milk yield in crossbred cattleSM Usman0https://orcid.org/0000-0001-8506-2929A Kumar1https://orcid.org/0000-0002-6495-7253A Sherasiya2https://orcid.org/0000-0002-1598-1820IM Youssef3https://orcid.org/0000-0003-3280-4495MM Abo Ghanima4https://orcid.org/0000-0002-1788-5082J Chandrakar5https://orcid.org/0009-0009-9596-1667R Tiwari6https://orcid.org/0000-0003-4426-7359NP Singh7https://orcid.org/0000-0002-8043-6298QS Sahib8https://orcid.org/0000-0003-4558-2860T Dutt9https://orcid.org/0000-0002-6029-0236AA Swelum10https://orcid.org/0000-0003-3247-5898Indian Veterinary Research InstituteIndian Veterinary Research InstituteStar Gulshan Park, NH-8A, Chandrapur Road, Wankaner, Gujarat, IndiaAgricultural Research CenterDamanhour UniversityIndian Veterinary Research InstituteIndian Veterinary Research InstituteIndian Veterinary Research InstituteSher-e-Kashmir University of Agricultural Sciences and Technology of KashmirIndian Veterinary Research InstituteKing Saud University This study was conducted using the first-lactation records of 1092 Vrindavani crossbred cattle to compare the relative efficiency of an artificial neural network (ANN) versus multiple linear regression for predicting the first-lactation 305-day milk yield (FL305DMY). The two input sets used for predicting FL305DMY in the study were input set-1: first four monthly test-day milk yields, age at first calving, and peak milk yield; and input set-2: first four monthly milk yields, age at first calving, and peak milk yield. The ANN was trained using a backpropagation algorithm based on Bayesian regularisation, and the algorithm was tested using four sets of training and test data at ratios of 66.67:33.33, 75:25, 80:20, and 90:10. The results revealed that the coefficient of determination showed no regular trend with decreasing the test dataset. Nevertheless, the observed values were highest for the 90:10 ratio of training-test data for both input sets, with the lowest root mean square error. The ANN model outperformed the multiple linear regression model when predicting FL305DMY, with an accuracy of 79.09% for input set-1 and 83.67% for input set-2, with the lowest root mean square error values for both input sets. Therefore, the ANN model can be used as an alternative technique to predict FL305DMY in Vrindavani cows. Submitted 30 June 2024; Accepted 20 December 2024; Published January 2025 ------------------------------------------------------------------ Significance of research to South African science The research demonstrates the potential of artificial neural networks (ANNs) to enhance dairy herd management through accurate early prediction of first-lactation 305-day milk yield in crossbred cattle. The study compares ANN and multiple linear regression models, showing that ANNs provide superior predictive accuracy and lower error margins. This has important implications for local dairy farming systems, particularly in data-limited environments, as it offers a practical tool for early identification of high- and low-producing cows. By enabling data-driven selection and culling decisions, the approach supports improved productivity, genetic progress, and economic efficiency in the South African livestock sector. https://www.sajas.co.za/article/view/23007Bayesian regularisationmilkmultiple linear regressionVrindavani cattle
spellingShingle SM Usman
A Kumar
A Sherasiya
IM Youssef
MM Abo Ghanima
J Chandrakar
R Tiwari
NP Singh
QS Sahib
T Dutt
AA Swelum
Artificial neural networks for predicting first-lactation 305-day milk yield in crossbred cattle
South African Journal of Animal Science
Bayesian regularisation
milk
multiple linear regression
Vrindavani cattle
title Artificial neural networks for predicting first-lactation 305-day milk yield in crossbred cattle
title_full Artificial neural networks for predicting first-lactation 305-day milk yield in crossbred cattle
title_fullStr Artificial neural networks for predicting first-lactation 305-day milk yield in crossbred cattle
title_full_unstemmed Artificial neural networks for predicting first-lactation 305-day milk yield in crossbred cattle
title_short Artificial neural networks for predicting first-lactation 305-day milk yield in crossbred cattle
title_sort artificial neural networks for predicting first lactation 305 day milk yield in crossbred cattle
topic Bayesian regularisation
milk
multiple linear regression
Vrindavani cattle
url https://www.sajas.co.za/article/view/23007
work_keys_str_mv AT smusman artificialneuralnetworksforpredictingfirstlactation305daymilkyieldincrossbredcattle
AT akumar artificialneuralnetworksforpredictingfirstlactation305daymilkyieldincrossbredcattle
AT asherasiya artificialneuralnetworksforpredictingfirstlactation305daymilkyieldincrossbredcattle
AT imyoussef artificialneuralnetworksforpredictingfirstlactation305daymilkyieldincrossbredcattle
AT mmaboghanima artificialneuralnetworksforpredictingfirstlactation305daymilkyieldincrossbredcattle
AT jchandrakar artificialneuralnetworksforpredictingfirstlactation305daymilkyieldincrossbredcattle
AT rtiwari artificialneuralnetworksforpredictingfirstlactation305daymilkyieldincrossbredcattle
AT npsingh artificialneuralnetworksforpredictingfirstlactation305daymilkyieldincrossbredcattle
AT qssahib artificialneuralnetworksforpredictingfirstlactation305daymilkyieldincrossbredcattle
AT tdutt artificialneuralnetworksforpredictingfirstlactation305daymilkyieldincrossbredcattle
AT aaswelum artificialneuralnetworksforpredictingfirstlactation305daymilkyieldincrossbredcattle