Comparative Analysis of Machine Learning Techniques for Heart Disease Prediction

Due to its substantial effects on society, efforts to enhance heart disease diagnosis and treatment have increased. Through data mining and the archiving of medical records, improved patient management is now possible thanks to the fusion of technology and medical diagnoses. Understanding the interaction between risk factors in the histories of patients and their impact on their prospects for heart disease is of utmost importance. This study aims to accurately predict cardiac disease by analyzing various data elements from patients. The selection and feature extraction methods are the most efficient choices for prediction system for heart disease. The following variables, such as age, sex, occupation, smoking, obesity, diet, exercise, mental stress levels, type of chest pain, history of chest pain, pressure, ECG, and results, have all been identified as significant factors in diagnosing heart disease. Various machine learning methods, such as Multilayer Perceptron (MLP), Naive Bayes (NB), K-nearest Neighbor (K-NN) and Support Vector Machine (SVM) were utilized to analyze the heart disease dataset. These datasets consisted of one with all features and another with only selected features. The objective was to compare the performance of these methods and determine which ones yielded accurate predictions. The results revealed that random forest, using the chosen features, outperformed other artificial intelligence algorithms, including those using all input features, achieving a maximum accuracy rate of 90%. As a support framework for predicting cardiac disease in its early stages, the suggested approach shows promise. This study advances the prognosis of cardiac disease, enables prompt therapies, and enhances patient outcomes by fusing the strength of data-driven AI algorithms with thorough feature selection.