Heart Disease Prediction Using Machine Learning

Abstract

This paper revolves around a classification use case of machine learning in which the intention is to predict the possibility of a heart disease in an individual given certain parameters. Machine Learning is extensively being used across the world. The healthcare industry has also commenced leveraging these data driven techniques. Machine Learning can play a vital role in predicting the likelihood of locomotor disorders, Heart ailments and more such diseases because machine learning is well known for its use cases in classifying, categorizing and predicting. Such information, if predicted well, can provide key foresight to doctors who can hence mould their diagnosis and course of treatment per patient basis. The main advantage of using machine learning in healthcare is its ability to parse and process huge datasets which are beyond the scope of human abilities, and then accurately convert the derived analysis of that data into clinical insights that can aid medical practitioners round the globe in planning stratergies for providing care to patients, ultimately leading to more promising results, reduced costs of care and last but not the least , increased patient satiation and response/recovery. To simplify and solve this problem, solutions were provided using multiple supervised learning algorithms like logistic regression, Naïve Bayes, random forests, decision trees, support vector machines and K-nearest neighbours. The best accuracy was seen using random forests.

Ähnliche Arbeiten