Abstract:
throughout the body. Heart health must be maintained because the mortality rate caused
by heart disease is among the highest in the world. So early action is needed to predict
the diagnosis of heart disease as a form of prevention or treatment efforts so that there is
no increase in cases of heart disease. Therefore, a system is needed that can help predict
the diagnosis precisely and accurately and on time based on a computer. This study aims
to build a prediction model for the diagnosis of heart disease using the Random Forest
method. This prediction model uses the RSI Ibnu Sina medical record dataset of 336 data,
including 268 used as training data and 68 testing data. The data attributes used were 11
attributes, namely gender, age, chest pain, systolic blood pressure, diastolic blood
pressure, cholesterol, current blood sugar (GDS), RestingECG, heart rate, ST Slope, and
diagnosis. This study produced a prediction model with an accuracy of 85.29% measured
using the Confusion Matrix.
