Open AccessPredicting the body weight of Hereford cows using machine learning
Author(s) -
Alexey Ruchay,
В. И. Колпаков,
Vsevolod Kalschikov,
К. М. Джуламанов,
Konstantin Dorofeev
Publication year - 2021
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/624/1/012056
Subject(s) - withers , girth (graph theory) , mathematics , coefficient of determination , mean absolute error , mean squared error , algorithm , statistics , body weight , linear regression , machine learning , correlation coefficient , artificial intelligence , computer science , medicine , combinatorics
Various machine learning algorithms have been used to model and predict the body weight of Hereford cows. The traditional linear regression model and various machine learning algorithms have been used to develop models for the prediction of the body weight of Hereford cows. The dependent variables include body weight and independent variables include withers height, hip height, chest dept, chest width, width in maclocks, sciatic hill width, oblique length of the body, oblique rear length, chest girth, metacarpus girth, backside half-girth, and age measurements of 1500 cows aged 2–6 years of age. The performance of the models is assessed based on evaluation criteria of the coefficient of determination, the root mean squared error, the mean absolute error, the mean absolute percentage error. We used a concept of splitting data into training, testing and validation datasets to provide a robust method for modelling and predicting. The RandomForestRegressor algorithm was found to provide the best results for training and testing datasets. It was concluded that machine learning algorithms may provide better results than the traditional models and may help researchers choose the best predictors for body weight of animals.