Heart Disease Prediction Model Based on Model Ensemble

EasyChair Preprint 3046

Abstract

With the increase of heart disease patients in the current society, early prevention of the heart health has become the top priority. The paper proposed a new heart disease prediction method based on model ensemble, which combined three independent models including SVM, decision tree and ANN. As for the data source, the public dataset published by UCI has been preprocessed and used into the ensemble model and three individual models. The prediction performance has been evaluated by accuracy, precision, recall, and f1 score indicators. As the result shown, compared with three independent models, the ensemble model has a better performance generally indicating a practical medical use.

Keyphrases: Decision Tree, Heart Disease Prediction, model ensemble

@booklet{EasyChair:3046,
  author    = {Xu Wenxin},
  title     = {Heart Disease Prediction Model Based on Model Ensemble},
  howpublished = {EasyChair Preprint 3046},
  year      = {EasyChair, 2020}}