Enhancing Heart Disease Prediction Accuracy: A Comparative Study of Machine Learning Models with Ensemble Method

Computer Science and Engineering

ABSTRACT Heart disease remains a critical global health concern, driving mortality rates and presenting challenges for early detection and treatment. Leveraging modern medical advancements, our study employs a multifaceted approach integrating electronic health records and online-connected regulators with wearable medical sensors. We utilize data mining techniques to efficiently process the continuous stream of human-generated health data, focusing on accurate classification for early heart disease detection. Our methodology encompasses meticulous data pre-processing, including missing value imputation, normalization, and categorical feature encoding. We employ a diverse array of machine learning algorithms, ranging from traditional logistic regression to advanced methods like random forests and support vector machines, optimizing them through rigorous experimentation and hyper-parameter tuning. Crucially, we emphasize feature selection to identify the most influential predictors of heart disease risk. Evaluation metrics such as accuracy, precision, recall, F1 score, and AUC-ROC underscore the effectiveness of our models, highlighting significant performance advantages for certain algorithms.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.