Beyond Boundaries: Achieving 100% Heart Disease Prediction Using Diverse Machine Learning Algorithms

Abstract

Heart disease plays a virtual role in recent years.This study addresses critical need for early detection of heart disease to alleviate its impact .We propose a machine learning architecture for early stage of the art feature extraction utilizing states of the art feature extraction stratagies. The effect of 5 machine learning algorithm-logistic algorithm - k-neighbour - support vector machine - decision tree - random forest was evaluated. Among the 5 algorithm random forest exhibited superior performance .We used 2 heart disease datasets -one 303 instances and another with 1026 instances the larger datasets yields outstanding results with the random forest achieving perfect scores across all metrics - accuracy-1,precision-1,and recall-1 and f1-score-1.This impressive performance underscores the algorithm’s effectiveness. The main objective of the project is to resolve and address specific issues