COMPARATIVE MACHINE LEARNING ALGORITHM FOR CARDIOVASCULAR DISEASE PREDICTION

Abstract

In the present study, It used to categorise heart illness in the Cleveland UCI repository. It visually describes the dataset, operational parameters, and predictive analytics development. Machine learning (ML) begins with data preparation. The technique uses an ML model and key parameters to predict cardiovascular illness in patients. The dataset comprises 14 heart disease characteristics for this investigation. The preliminary examination and evaluation predicted heart problems. The dataset has 303 samples with 14 features. The information is presented as a percentage of truth. KNN 86%, Decision Trees 79%, Logistic Regression 85%, Naive Bayes 86%, and Support Vector Machines 87% can predict heart disease 89% accurately. The receiver working characteristics show that the random forest technique for heart disease prediction has an 89% diagnostic rate. The proposed method uses the random forest algorithm since it has been shown to be the most effective algorithm for classifying cardiovascular illness.

Keywords

• heart disease prediction
• parameters
• machine learning
• random forest
• decision tree