Congestive heart failure or Congestive Heart Failure (CHF) is the number one cause of death in the world. There are approximately 5.7 million adults with heart failure in the United States and 550,000 new cases are diagnosed each year. This has encouraged a lot of research on heart failure, one of which is using the Machine Learning method to predict death from heart failure early. From these problems, the authors will conduct Machine Learning research using two different algorithm models, namely K-Nearest Neighbor (KNN) and Support Vector Machine (SVM). These two models will predict death due to heart failure. The dataset regarding the factors for diagnosing heart failure can be accessed widely and freely on the Kaggle website which is divided into two, namely data training and data testing then analysis and prediction are carried out, so that information is obtained in the form of an accuracy rate in predicting death in heart failure. Using this function also produces the accuracy of each model on the data that has been trained. Data taken were 299 patient data with 13 features or attributes, then divided into 239 training data and 60 test data. The value obtained is an accuracy of 85%. The accuracy obtained is more than 80% of the total data tested so that it can be used or implemented to classify heart failure.
Copyrights © 2022