A comparison of neural network and multiple regression predictions for 305-day lactation yield using partial lactation records

Abstract

A comparison of neural network and multiple regression predictions for 305-day lactation yield using partial lactation records. Can. J. Anim. Sci. 83: 307-310. Milk yield predictions based on artificial neural etworks and multiple regression were studied. The 305-d lactation yield predictions were based on milk yield of the first 4 test days. Average 305-d milk production of the herd, number of days in milk and month of calving. The predictions made with either the neural network or the multiple regression model did not differ (P > 0.05) from the values estimated with the current Polish dairy cattle evaluation system. The neural network model may be alternative method of predicting these traits.Key words: Artificial neural networks, multiple linear regression, milk yield prediction, test day data Grzesiak, W., Lacroix, R., Wójcik, J. and B⁄ laszczyk, P. 2003. Comparaison de réseaux de neurones artificiels et de régression multiple pour la prédiction de rendements laiters à 305 jours à partir de données partielles de lactation. Can. J. Anim. Sci. 83: 307-310. L'utilisation de réseaux de neurones artificiels et de la régression multiple pour la prédiction des rendements laitiers a été étudiée. Les prédictions des rendements laitiers à 305 jours était basées sur le lait produit au quatre premiers tests, la moyenne de production laitière à 305 jours du troupeau, le nombre de jours en lait et le mois du vêlage. Les prédictions faites par le réseau de neurones et le modèle généré par régression multiple ne différaient pas de façon significative des valeurs estimées par le système polonais actuellement utilisé pour l'évaluation des troupeaux laitiers. Les réseaux de neurones artificiels constituent donc une alternative pour le développement de modèles prédictifs.Mots clés: Réseaux de neurones artificiels, régression linéaire multiple, prédiction des rendements laitiers, données au jour du testIn many countries, analysis of milk yield for 305-d lactation represents a foundation for dairy cattle genetic evaluation. Various regression models, which are used for future yield predictions, are applied in milk yield analyses. In comparison with regression methods or time-series analyses, artificial neural networks (ANN) represent a different approach. These solve particular problems through learning, by typical inputs and respective desired responses, unlike conventional methods that consist of the construction of an algorithm and its implementation as a computer programme (Tadeusiewicz 1993). In practice, ANNs have primarily been used in engineering, economic predictions, or in medical diagnoses. Recently, they have also been used in some areas of animal genetics and husbandry, such as swine production environment control, or piglet behaviour analysis. ANNs may find application in food processing, e.g. for fermentation control. Yang et al. (1999) applied ANNs to analyses related to predicting clinical mastitis in cattle and found that the technology was able to determine major factors related to the presence or absence of m...

Prediction of 305-day milk yield in Brown Swiss cattle using artificial neural networks

“…However, 305-d milk yield prediction by MLR was lower than the observed 305-d milk yield by 629.4 kg (P <0.01). Grzesiak et al (2003) used ANN and MLR in their study, showing there was no significant difference between observed values and predicted values (P >0.05) suggesting that ANN was appropriate for modelling 305-d milk yield. The average 305-d milk yield predicted by the ANN was lower than the average observed yield of the 49 reference cows by 13.12 kg ( Table 1).…”

“…y was the actual value, n was the total number of records (Salehi et al, 1998;Grzesiak et al, 2003). Ratio of mean (RoM) described by Friedrich et al (2008) was calculated as:…”

“…Kominakis et al (2002) tested the usefulness of ANNs in predicting lactation, as well as daily test milk yield(s) in Chios dairy sheep based on a few (2 -4) test-day records in the beginning of a lactation period. Grzesiak et al (2003) compared the neural network and multiple regression predictions for 305-d lactation yield using partial lactation records. Sharma et al (2007) used an ANN model to predict the first lactation 305-day milk yield using partial lactation records pertaining to Karan Fries crossbred dairy cattle.…”

Qualitative Analysis of Feed Management Practice on Goat Herds by Self Organizing Maps in Murcia Region of Spain

“…Grzesiak et al (2003) Data were analyzed using a neural network, the so-called Self-organizing Map, SOM (Kohonen, 1997). This network was formed by elementary units, neurons, in a bidimensional network.…”

Examples of the Use of Data Mining Methods in Animal Breeding

“…A similar study on the use of ANNs for regression problems concerned predictions for 305-day lactation yield in Polish Holstein-Friesian cows based on monthly test-day results [48]. The following 7 input variables were used: mean 305-day milk yield of the barns in which the cows were utilized, days in milk, mean test-day milk yield in the first, second, third and fourth month of the research period and calving month.…”