<![CDATA[Heart failure is a serious and pressing health problem that affects millions of people worldwide. Several factors influence the occurrence of heart failure, such as age, type of pain, blood pressure, cholesterol levels, and other risk factors associated with heart disease. With current technological developments, data mining and machine learning can be used to predict patient health conditions. Therefore, the problem of this research is how to implement data mining techniques for identifying heart disease. The goal of the study is to identify heart disease and prevent heart failure. This study utilises the K-Nearest Neighbour (k-NN) algorithm to estimate the likelihood of patients experiencing heart failure based on available data features. The data used is taken from the kaggle.com site, which includes information from patients diagnosed with heart failure and those who do not suffer from heart failure. The analysis process involves data processing steps, such as normalisation, feature grouping, and selecting the optimal K parameter for the k-NN algorithm. Evaluation is carried out by calculating the accuracy, precision, recall, and F1-score values. Testing is carried out on a dataset with 299 patient data, which is divided into training data and test data with a ratio of 80:20. The results of this study indicate that the k-NN algorithm has an accuracy of 87% in predicting kidney failure. This result indicates that the k-Nearest Neighbour algorithm can effectively predict heart failure.]]>
