Document type: meta-analysis published in Journal of Veterinary Behavior
Authors: Lianlian Fu, Yu Le
Preview: Over the past few decades, animal welfare has become an increasingly prominent focus across various sectors of the livestock industry. Among them, pig welfare has garnered global attention due to its profound implications for animal health, productivity, and public trust, as well as growing consumer willingness to purchase higher-welfare pork products. This study presents a systematic review of the scientific literature on pig welfare using text mining (TM) techniques-an analytical approach that remains underutilized in this field. A total of 7,031 articles were retrieved from the Scopus® database (1991-2024), of which 1,331 met strict inclusion criteria. The study employed TM methods, including Term Frequency-Inverse Document Frequency (TF-IDF) and Latent Dirichlet Allocation (LDA) modeling, to identify prevailing topics and thematic trends. Seven major topics were extracted: stress and treatment, social behavior, transport and slaughter, housing and welfare, health and disease assessment, feeding and reproduction, and farm management. The academic output in this field has shown exponential growth, with the majority of publications originating from Europe. Keywords such as " tail, " " transport, " and " wean " emerged as focal points of interest. These findings highlight the potential of TM in constructing domain knowledge and identifying research gaps, offering valuable insights for scholars, practitioners, and policymakers committed to improving animal welfare systems.
