A K-Nearest Neighbor (KNN)-Algorithm in Poultry Diseases Monitoring System

Abstract

In poultry farming, the most important factor during their growth and production is for the birds to be free from any disease. Unmonitored and uncontrolled temperature and humidity within the cage can lead to reduced productivity and high rate of mortality. Diseases such as: avian influenza, Newcastle, fowl paralysis, semolina and so on may wipe up an entire farm in few minutes. This research used K- nearest neighbor (KNN) algorithm and developed a model to monitor such factors automatically. In the model, digital camera was deployed in the cage that records the behavior of the fowls. The model was trained on the normal behavior of a healthy/unhealthy fowl based on the clinical signs so that any abnormal sign reported immediately by the system. The experimental results show that the model can report the various diseases sign in poultry farms.