PREDICTIVE MODELING OF CARDIOVASCULAR DISEASE US-ING MACHINE LEARNING

Abstract

The rising incidence of cardiac conditions is becoming a significant concern; therefore, it is crucial to anticipate these cases in advance. Consequently, arriving at this diagnosis presents a challenging endeavor that must be executed swiftly and with precision. The primary objective of this study is to assess an individual's likelihood of developing a cardiac condition by considering various medical factors. A tool was developed to predict the likelihood of a heart disease diagnosis, utilizing the patient's medical history as a basis for its predictions. We utilized several machine learning methods, including logistic regression and KNN, to forecast and categorize individuals diagnosed with heart disease. A systematic approach was utilized to regulate the model's implementation, aiming to enhance the precision of predicting heart attacks in individuals. The model demonstrated a significant level of precision in forecasting indicators of heart disease in a person through the application of KNN and Logistic Regression techniques. The accuracy achieved surpassed that of previous classifiers, such as naive Bayes. The implementation of the proposed model has notably alleviated stress by diminishing the chances of the classifier accurately and precisely detecting cardiac conditions. The proposed method for predicting cardiovascular disease reduces expenses and improves healthcare quality. This project collects substantial data that may aid in predicting individuals at risk of developing heart disease. Pynb files serve as the medium for its implementation.