Unsupervised and supervised machine learning approach to assess user readiness levels for precision livestock farming technology adoption in the pig and poultry industries.

This study used machine learning, particularly k-means clustering, to identify distinct clusters of users and their technological readiness to adopt various precision livestock farming (PLF) technologies based on their responses to a carefully designed questionnaire. The analysis revealed initially two as well as three distinct clusters representing different levels of technological readiness among farmers considering the adoption of various PLF technologies. In addition to the validation of the cluster results by internal metrics, a related principal component analysis, and a focus group evaluation, this paper describes the application of a Decision Tree as an explainable supervised machine learning approach to investigate the predictive power of specific survey questions. In combination, this research aims to provide valuable insights for understanding farmers' technological readiness, to enhance further survey designs, and to support the development of targeted strategies to promote the successful adoption of PLF technologies in the agricultural sector generally. • Technological readiness of pig and poultry farms was explored. • Two and three clusters of technological readiness were observed by K-Means. • User characteristics for targeted interventions have been highlighted. • Importance of survey questions for cluster identification was calculated through Entropy. • Results have been validated by internal metrics, a PCA, and a Focus Group. [ABSTRACT FROM AUTHOR]