PREDICTION OF FEED INTAKE AND BODY WEIGHT IN RABBITS USING MACHINE LEARNING

Abstract

Livestock management practices, particularly in the context of rabbit breeding, are crucial for ensuring optimal growth performance and productivity. By closely monitoring factors such as body weight and feed intake, breeders can make informed decisions to enhance the overall efficiency of their operations.This project work investigated the performance of different rabbit breeds in terms of weight gain and feed consumption over varying weeks in a controlled experimental setting. The study was conducted at the University of Benin Farm Project in Nigeria, utilizing data collected on body weight and feed intake on a weekly and daily basis respectively. Statistical analysis was carried out in R version 4.0.3. The results revealed breed-specific variations in weight gain and feed consumption, with breeds like Angora and Chinchilla exhibiting superior body weight efficiency while consuming comparable amounts of feed compared to Dutch and Hyla breeds respectively. Weekly performance analysis indicated fluctuations in weight gain across different weeks, with week 10 showing the highest growth performance. Additionally, cluster analyses identified optimal performing groups of animals based on specific criteria. Different machine learning algorithms was used in characterization of feed intake and weight gain for different rabbits breeds, such as Support Vector Machines (SVM), Random Forest (RF) regression, Decision regression and cluster analysis, to predict feed intake and body weight with varying levels of accuracy, with cluster analysis and decision tree proving to be the best and SVM the least. The findings highlight the potential of machine learning in automating feed intake and weight gain estimation and improving prediction accuracy in livestock management practices. Also, this research provides valuable insights into breed-specific growth patterns, optimal growth phases, and the potential of machine learning in enhancing feed intake predictions in rabbit farming. The results contribute to the existing knowledge on growth efficiency in rabbits and offer practical implications for improving breeding programs and management practices in the livestock industry.