In the livestock industry, the concerns of society and livestock farmers have changed in recent decades, moving from a focus solely on production traits to sustainable production while respecting animal welfare (Ducrot et al., 2024). Animal behavior parameters can serve as a reliable proxy for animal welfare, but collecting large data sets of behavioral data on farm is very time-consuming when not automatized.  Accelerometers are promising devices to detect animal behavior. A key element of the efficiency of the prediction of animal behavior from accelerometer data is the adequacy of pre-processing methods (Riaboff et al., 2019,Riaboff et al., 2022, Vidal et al., 2023).  

The article of Mauny et al., (2025) aims to find a solution for using huge automatized data. So, livestock farmers can effectively use new technologies to monitor animal behavior and then correct poor husbandry routines. Authors used the pipeline ACT4Behav - (Accelerometer-based Classification Tool for identifying Behaviours) - (Mauny et al., 2024) for the pre-processing of accelerometer data with the aim of selecting the most important features of dairy goats related to rumination, head in the feeder, lying and standing (using  data based on video recordings for the validation of behavior activity). Mauny et al., (2025) established a clear methodology that systematically obtain the best features and processing techniques to predict each targeted variable from raw accelerometer data. The work provides valuable information for both animal husbandry specialists and data mining scientists. The main limitation of the work is the small number of animals for both training and model testing. 

References 

Ducrot C, Barrio MB, Boissy A, Charrier F, Even S, Mormède P, Petit S, Pinard-van der laan M-H, Schelcher F, Casabianca F, Ducos A, Foucras G, Guatteo R, Peyraud J-L, Vayssier-Taussat M, Veysset P, Friggens NC and Fernandez X. 2024. Animal board invited review: Improving animal health and welfare in the transition of livestock farming systems: Towards social acceptability and sustainability. animal 18, 101100. https://doi.org/10.1016/j.animal.2024.101100

Mauny, S, Kwon J, Friggens NC, Duvaux-Ponter C, Taghipoor M. 2024. ACT4Behav (Accelerometer-based Classification Tool for identifying Behaviours): a Machine Learning pipeline with extensive pre-processing and feature creation options. Zenodo. https://doi.org/10.5281/zenodo.12624796

Mauny S, Kwon J, Friggens NC, Duvaux-Ponter C, Taghipoor M. 2025. A pipeline with pre-processing options to detect behaviour from accelerometer data using Machine Learning tested on dairy goats.. Zenodo, ver. 6 peer-reviewed and recommended by PCI Animal Science. https://doi.org/10.5281/zenodo.12627197

Riaboff L, Aubin S, Bédère N, Couvreur S, Madouasse A, Goumand E, Chauvin A, Plantier G. 2019. Evaluation of pre-processing methods for the prediction of cattle behaviour from accelerometer data. Computers and Electronics in Agriculture 165, 104961. https://doi.org/10.1016/j.compag.2019.104961

Riaboff L, Shalloo L, Smeaton AF, Couvreur S, Madouasse A, Keane MT. 2022. Predicting livestock behaviour using accelerometers: A systematic review of processing techniques for ruminant behaviour prediction from raw accelerometer data. Computers and Electronics in Agriculture 192, 106610. https://doi.org/10.1016/j.compag.2021.106610

Vidal G, Sharpnack J, Pinedo P, Tsai IC, Lee AR, Martínez-López B. 2023. Impact of sensor data pre-processing strategies and selection of machine learning algorithm on the prediction of metritis events in dairy cattle. Preventive Veterinary Medicine 215, 105903. https://doi.org/10.1016/j.prevetmed.2023.105903