Journal of Emerging Trends in Engineering and Applied Sciences: Research Areas

Abstract:
Poultry diseases is a serious and growing problem with increased fatality rate. Early detection of poultry diseases is the first step in eradicating and preventing its spread as well as improve the health of the birds and reduce mortality rate. In this paper we proposed a Bayesian belief network (BBN) model for predicting fourteen different types of poultry diseases. The BBN model had 41 nodes which contains various poultry disease symptoms and diseases. The BBN model was trained using the likelihood sampling algorithm and an average prediction accuracy of 89.4% on the fourteen different types of poultry diseases was achieved. The proposed model will assist farmers, especially in the rural area in diagnosing poultry diseases and this will help in reducing mortality rate in sick birds, as early detection of poultry diseases is the first step in eradicating and preventing its spread.

Keywords: poultry diseases, bayesian belief network, predicting.

Download full paper