Implementasi Algoritma Random Forest Dalam Klasifikasi Diagnosis Penyakit Stroke

Abstract

The most common disease in Indonesia is stroke, this disease occurs when blood flow to the brain is disrupted, either due to rupture of blood vessels or due to blockage of blood vessels. The data mining process can be a solution in identifying early symptoms of stroke. By using the Random Forest Method, it is hoped that it can be the right choice for preprocessing data in identifying early symptoms. The model results produce an adjustment of 96% of the training score and from the results table of precision, recall, F1-score, and accuracy which results in an accuracy of 0.95 or 95%, as well as the final result of AUC of 0.80 which shows that the model results are included in the good classification