Text mining approach in chicken meat welfare.

Area of Science:

• Animal Science
• Agricultural Science
• Data Science

Background:

• Broiler welfare research has seen a surge in publications since the 1990s, with a notable increase in the last decade attributed to public interest.
• The USA, Brazil, and Europe-27 are leading contributors, publishing 60% of the identified documents.

Purpose of the Study:

• To analyze factors affecting broiler welfare using Text Mining techniques.
• To identify relationships between different elements influencing broiler welfare through data analysis.

Main Methods:

• A comprehensive literature search was conducted on Scopus using terms related to "Welfare" and "Chicken" (including "Broiler", "Broiler chicken", "Chicken meat").
• Text Mining and Topic Analysis were applied to 1228 abstracts using R software (4.3.1), analyzing word frequency, bigram, and trigram weights.
• Limitations of Text Mining, such as ambiguity and context sensitivity, were acknowledged.

Main Results:

• Key terms identified include slaughter phase, housing management, and environmental conditions (light quality/quantity).
• A strong correlation was found between rearing, slaughter phases, litter management, and broiler welfare.
• Research topics vary by region, focusing on genetic selection, feeding, stocking density, slaughter, and consumer perceptions.

Conclusions:

• The substantial increase in publications underscores the global importance of broiler welfare.
• Despite geographic variations, common research themes are challenging to pinpoint, indicating diverse research priorities over time and across regions.