CASE-BASED REASONING UNTUK DIAGNOSA PENYAKIT THT (TELINGA HIDUNG TENGGOROK); CASE-BASED REASONING FOR DIAGNOSIS ENT DESEASES (EAR NOSE THROAT)

TEDY RISMAWAN, Sri Hartati

2012 | Disertasi | PROGRAM STUDI S2 ILMU KOMPUTER

Case-Based Reasoning (CBR) is a reasoning system that uses old knowledge to solve new problem. CBR provides solutions to new cases by looking at an old case that comes closest to the new case. It will be very useful because it eliminates the need to extract the model as required by the rule-based systems. Moreover, CBR can also be started from a small amount of knowledge, because the knowledge of CBR can be increased gradually when a case is added. This study tries to establish a system for Case-Based Reasoning for diagnosing diseases of ENT (Ear, Nose, Throat) by looking at the characteristics of existing symptoms in patients. Diagnosis process is done by inserting a new case (target case) that contains the symptoms of the disease to be diagnosed into the system, then the system will do the indexing process with backpropagation method to obtain an index of new cases. After obtaining an index of the case, then the system do the calculation of the value of similarity between the new case by case basis which has the same index with new cases using cosine coefficient method. The case taken is the case with the highest similarity value. If a case is not successfully diagnosed, then the case will be revised by the experts. Revised successful cases will be stored into the system to be used as new knowledge (fresh knowledge) for the system. The results showed case-based reasoning system for diagnosing diseases of ENT can help paramedics in performing diagnostics. The test results of 111 data test cases, obtained 9 cases that have similarity values below 0.8

Kata Kunci : THT, penalaran berbasis kasus; indexing; backpropagation; similarity