Classification of Swine Disease Using K-Nearest Neighbor Algorithm on Cloud-Based Framework

A total of 25.9% of global jobs are occupied by the agriculture industry. As the global population grows, demand for both livestock and food production is growing rapidly. Many farmers live in rural parts of the nation and are in total doubt as to whether or not their livestock and crops are healthy. As a precautionary measure, they contact veterinary physicians to stop the transmission of illnesses to other healthy animals, which is counterproductive to their palatable fitness. By offering a cloud-based method to decide whether animals are safe or unsafe, this research aims to solve this issue. This research also aims to use supervised learning to conduct real-time classification, along with elucidating the device architecture that classifies photographs of diseased animals using the k-nearest neighbor (KNN) algorithm. A web cloud-based expert system was developed for the detection of swine diseases in pigs. MySQL was used for the cloud-based database which is at the backend of the website developed. PHP was used for the coding aspect of the system; the KNN algorithm was embedded into the PHP version 7.1.30. The developed system was able to predict disease that detects the likely occurrence of swine disease using ML algorithm. The system will allow farmers to input the symptoms noticed in the sick pigs, derive correct symptoms from a pig suffering a particular illness, and lastly prevent swine diseases from affecting other healthy pigs on the farm.