Abstract

Lung disease is one of the diseases with the highest number of patients in Indonesia. Lung disease is a disease with many types and symptoms that are almost the same as each other. This study uses an artificial neural network Learning Vector Quantization 3 (LVQ3), to diagnose lung disease. The data used in this study were 113 medical records, with seven types of lung disease, and 27 symptoms of the disease. From the experimental results, the best LVQ3 parameters from this study are using m = 0.15, and the learning rate = 0.15. LVQ3 produces the best accuracy value for training data at 87.5% of 80 data, and accuracy for test data 88% of 33 data.