Towards Precision Cardiology through Computational Modeling: Comparative Analysis of ML-Based Predictive Models

Abstract

Heart issues are one of the major and serious health issues that are affecting the mortality rate at worldwide level. Having heart-related problems can also increase the chances of having other minor and major health issues. Heart issues do not depend on age and gender factors, and they can occur at any age. In the age of advanced technology, such as ML, DL, AI, NLP, LLM and so on, are widely used in various domains such as banking, marketing, education, healthcare and so on. We used supervised learning to predict any diseases in their early stage because we have values with specific parameters and features in medical data. In this study, we used various machine learning methodologies such as Linear and Logistic Regression, DT, Random Forest, KNN, Support Vector Classification and Gradient Boosting. This study focused on a comparative analysis of heart failure chances using supervised machine learning. In future studies, this work can improve the quality of results more robustly by decreasing false positive values using advanced methods, such as deep learning, NLP, and BERT methods.