Revealing the Hidden Social Structure of Pigs with AI-Assisted Automated Monitoring Data and Social Network Analysis

Background: The social interactions of farm animals affect their performance, health and welfare. This proof-of-concept study addresses, for the first time, the hypothesis that applying social network analysis (SNA) on AI-automated monitoring data could potentially facilitate the analysis of social structures of farm animals. Methods: Data were collected using automated recording systems that captured 2D-camera images and videos of pigs in six pens (16-19 animals each) on a PIC breeding company farm (USA). The system provided real-time data, including ear-tag readings, elapsed time, posture (standing, lying, sitting), and XY coordinates of the shoulder and rump for each pig. Weighted SNA was performed, based on the proximity of "standing" animals, for two 3-day period-the early (first month after mixing) and the later period (60 days post-mixing). Results: Group-level degree, betweenness, and closeness centralization showed a significant increase from the early-growing period to the later one (p < 0.02), highlighting the pigs' social dynamics over time. Individual SNA traits were stable over these periods, except for the closeness centrality and clustering coefficient, which significantly increased (p < 0.00001). Conclusions: This study demonstrates that combining AI-assisted monitoring technologies with SNA offers a novel approach that can help farmers and breeders in optimizing on-farm management, breeding and welfare practices.