Predicting mastitis in dairy cows using neural networks and generalized additive models: a comparison

Publikation: Bidrag til tidsskrift/Konferencebidrag i tidsskrift /Bidrag til avisTidsskriftartikelForskningpeer review

Standard

Predicting mastitis in dairy cows using neural networks and generalized additive models : a comparison. / Anantharama Ankinakatte, Smitha; Norberg, Elise; Løvendahl, Peter; Edwards, David; Højsgaard, Søren.

I: Computers and Electronics in Agriculture, Bind 99, 11.2013, s. 1-6.

Publikation: Bidrag til tidsskrift/Konferencebidrag i tidsskrift /Bidrag til avisTidsskriftartikelForskningpeer review

Harvard

Anantharama Ankinakatte, S, Norberg, E, Løvendahl, P, Edwards, D & Højsgaard, S 2013, 'Predicting mastitis in dairy cows using neural networks and generalized additive models: a comparison', Computers and Electronics in Agriculture, bind 99, s. 1-6. https://doi.org/10.1016/j.compag.2013.08.024

APA

Anantharama Ankinakatte, S., Norberg, E., Løvendahl, P., Edwards, D., & Højsgaard, S. (2013). Predicting mastitis in dairy cows using neural networks and generalized additive models: a comparison. Computers and Electronics in Agriculture, 99, 1-6. https://doi.org/10.1016/j.compag.2013.08.024

CBE

MLA

Vancouver

Author

Anantharama Ankinakatte, Smitha ; Norberg, Elise ; Løvendahl, Peter ; Edwards, David ; Højsgaard, Søren. / Predicting mastitis in dairy cows using neural networks and generalized additive models : a comparison. I: Computers and Electronics in Agriculture. 2013 ; Bind 99. s. 1-6.

Bibtex

@article{76c227628b21492c91740518f8c20e91,
title = "Predicting mastitis in dairy cows using neural networks and generalized additive models: a comparison",
abstract = "The aim of this paper is to develop and compare methods for early detection of oncoming mastitis with automated recorded data. The data were collected at the Danish Cattle Research Center (Tjele, Denmark). As indicators of mastitis, electrical conductivity (EC), somatic cell scores (SCS), lactate dehydrogenase (LDH), and milk yield are considered. Each indicator is decomposed into a long-term, smoothed component, and a short-term, residual component, in order to distinguish long-term trends from short-term departures from these trends. We also study whether it is useful to derive a latent variable that combines residual components into a score to improve the model. To develop and verify the model, the data are randomly divided into training and validation data sets. To predict the occurrence of mastitis, neural network models (NNs) and generalized additive models (GAMs) are developed using the training set. Their performance is evaluated on the validation data set in terms of sensitivity and specificity. Overall, the performance of NNs and GAMs is similar, with neither method appearing to be decisively superior. NNs appear to be marginally better for high specificities. NNs model results in better classification with all indicators, using individual residuals rather than factor scores. When SCS is excluded, GAMs shows better classification result when milk yield is also excluded. In conclusion, the study shows that NNs and GAMs are similar in their ability to detect mastitis, a sensitivity of almost 75{\%} observed for 80{\%} of fixed specificity. Including SCS in the models improves their predictive ⩾5{\%} ability",
keywords = "Mastitis, Online data, Modeling techniques",
author = "{Anantharama Ankinakatte}, Smitha and Elise Norberg and Peter L{\o}vendahl and David Edwards and S{\o}ren H{\o}jsgaard",
year = "2013",
month = "11",
doi = "10.1016/j.compag.2013.08.024",
language = "English",
volume = "99",
pages = "1--6",
journal = "Computers and Electronics in Agriculture",
issn = "0168-1699",
publisher = "Elsevier BV",

}

RIS

TY - JOUR

T1 - Predicting mastitis in dairy cows using neural networks and generalized additive models

T2 - a comparison

AU - Anantharama Ankinakatte, Smitha

AU - Norberg, Elise

AU - Løvendahl, Peter

AU - Edwards, David

AU - Højsgaard, Søren

PY - 2013/11

Y1 - 2013/11

N2 - The aim of this paper is to develop and compare methods for early detection of oncoming mastitis with automated recorded data. The data were collected at the Danish Cattle Research Center (Tjele, Denmark). As indicators of mastitis, electrical conductivity (EC), somatic cell scores (SCS), lactate dehydrogenase (LDH), and milk yield are considered. Each indicator is decomposed into a long-term, smoothed component, and a short-term, residual component, in order to distinguish long-term trends from short-term departures from these trends. We also study whether it is useful to derive a latent variable that combines residual components into a score to improve the model. To develop and verify the model, the data are randomly divided into training and validation data sets. To predict the occurrence of mastitis, neural network models (NNs) and generalized additive models (GAMs) are developed using the training set. Their performance is evaluated on the validation data set in terms of sensitivity and specificity. Overall, the performance of NNs and GAMs is similar, with neither method appearing to be decisively superior. NNs appear to be marginally better for high specificities. NNs model results in better classification with all indicators, using individual residuals rather than factor scores. When SCS is excluded, GAMs shows better classification result when milk yield is also excluded. In conclusion, the study shows that NNs and GAMs are similar in their ability to detect mastitis, a sensitivity of almost 75% observed for 80% of fixed specificity. Including SCS in the models improves their predictive ⩾5% ability

AB - The aim of this paper is to develop and compare methods for early detection of oncoming mastitis with automated recorded data. The data were collected at the Danish Cattle Research Center (Tjele, Denmark). As indicators of mastitis, electrical conductivity (EC), somatic cell scores (SCS), lactate dehydrogenase (LDH), and milk yield are considered. Each indicator is decomposed into a long-term, smoothed component, and a short-term, residual component, in order to distinguish long-term trends from short-term departures from these trends. We also study whether it is useful to derive a latent variable that combines residual components into a score to improve the model. To develop and verify the model, the data are randomly divided into training and validation data sets. To predict the occurrence of mastitis, neural network models (NNs) and generalized additive models (GAMs) are developed using the training set. Their performance is evaluated on the validation data set in terms of sensitivity and specificity. Overall, the performance of NNs and GAMs is similar, with neither method appearing to be decisively superior. NNs appear to be marginally better for high specificities. NNs model results in better classification with all indicators, using individual residuals rather than factor scores. When SCS is excluded, GAMs shows better classification result when milk yield is also excluded. In conclusion, the study shows that NNs and GAMs are similar in their ability to detect mastitis, a sensitivity of almost 75% observed for 80% of fixed specificity. Including SCS in the models improves their predictive ⩾5% ability

KW - Mastitis

KW - Online data

KW - Modeling techniques

U2 - 10.1016/j.compag.2013.08.024

DO - 10.1016/j.compag.2013.08.024

M3 - Journal article

VL - 99

SP - 1

EP - 6

JO - Computers and Electronics in Agriculture

JF - Computers and Electronics in Agriculture

SN - 0168-1699

ER -