فهرست نشریات علمی دسترسی آزاد ایران

Background and Objective: Cardiovascular disease is a significant contributor to illness and death worldwide. In recent years, there has been a significant rise in the incidence of cardiovascular diseases. Since heart disease can be fatal without obvious symptoms, it is sometimes referred to as the "silent killer". Numerous research studies have been carried out in data mining to predict and calculate the risks of cardiovascular diseases. The healthcare industry's extensive data collection has made machine learning a powerful tool for decision-making and prediction. Method: By using machine learning algorithms and classifying patient data, it is possible to diagnose this disease. Ensemble learning, by combining prominent features from multiple models, aids in achieving consensus in prediction. By combining the results of different models, prediction performance can be improved. Research has shown that more diverse statistical models yield better results in the diagnosis and prediction of cardiovascular disease. This article utilizes various algorithms such as Support Vector Machines, Naive Bayes, Gradient Boosting Machines, Random Forest, Decision Tree, K-Nearest Neighbors, Logistic Regression for predicting heart disease. Results The system employs weighted voting techniques to ensemble the models and cross-validation to evaluate our methods. Logistic Regression, Naive Bayes, and Gradient Boosting Machine provide the highest accuracy rates of 83.19%, 81.53%, and 79.56% respectively on this dataset. The weighted voting technique yields the best result with an accuracy of 87.78%. Conclusion: This article's main objective is to develop a voting technique that can accurately predict cardiovascular diseases in the real world.