Penerapan Case-Based Reasoning pada Sistem Cerdas untuk Pendeteksian dan Penanganan Dini Penyakit Sapi

Abstract

Cow disease causes significant impact on meat production degradation for livestock breeders. Independent early detection and treatment using intelligent system is necessary to minimalize the bad impact caused by cow disease. The former research has produced intelligent system based on Backpropagation Artificial Neural Network(ANN). But unfortunately ANN works in a blackbox because the groundwork information which lead to the diagnostic result is unshown. Intelligent system based on Cased-Based Reasoning(CBR) can solve the problem that happened in ANN based intelligent system because CBR suggests the diagnosis result based on past problems which can be revised to solve new problem accurately. Besides that, CBR can show the groundwork information from its diagnostic result. From the three experiments, using cases from case memory (scenario 1), outside case memory (scenario 2), and partial symptoms from case memory (scenario 3) gives good result with 100% and 95.83% precision value for scenario 1 and 3. And precision value which is actually true with only 59.31% for scenario 2. The poor result in scenario 2 caused by the cases used in experiment are cases that never been faced by intelligent system, so there is no cases in case memory has the same symptoms with the symptom entered in intelligent system. So it can be concluded from the experiments that this intelligent system can suggests diagnostic result accurately and eases the breeders to diagnose cow disease independently. Abstrak