Cardiovascular Disease Prediction Based on Soft Voting Ensemble Model

Abstract

In recent years, cardiovascular disease has become a serious threat to the health and safety of people all over the world. Machine learning, deep learning and other artificial intelligence (AI) technologies used to assist medical diagnosis are becoming more and more popular. In order to improve the performance of cardiovascular disease prediction, this paper proposes an ALD soft voting ensemble model (ALD-SVE), which is composed of three individual learners, Attentional Factorization Machines (AFM) can make full use of the cross features of cardiovascular disease data and capture. The attention mechanism introduced in the AFM model gives different weights to cross features to enhance interpretation, and then uses the soft voting ensemble of Logistic Regression (LR) and Decision Tree (DT) to further enhance the stability and generalization of the model. Experiments show that the ALD-SVE model has excellent performance in cardiovascular disease data sets, and its AUC value reached 0.80306, better than the listed comparison model.