BİLDİRİLER

BİLDİRİ DETAY

Ecem İREN
HEART DISEASE PREDICTION WITH MACHINE LEARNING ALGORITHMS
 
Heart diseases are among the main causes of death all over the world. According to the report of World Heart, more than half a billion in the world is affected by cardiovascular diseases. Unfortunately, 20.5 million death records belong to the year 2021. Also, it is mentioned that approximately 80% of the early detected cardiovascular disease such as heart attacks can be prevented therefore it is really significant to predict heart disease in advance to reduce the number of deaths. This is a critical subject that needs to be studied and taken effective precautions to reduce heart-related death rate. Diagnosing the disease accurately and in a timely manner is very important to achieve the goal. In this context, machine learning algorithms can play significant role at predicting the heart disease in early stages according to some parameters. In this study, heart disease prediction accuracies of machine learning models such as Logistics Regression, Support Vector Machine, K-Nearest Neighbors algorithm, Decision Tree, Naïve Bayes and Random Forest classifiers are compared. After the experiments with Cleveland dataset, it is observed that K-Nearest Neighbors classification algorithm gives the best accuracy result with 83.61% among the other models. On the contrary, the worst classification result belongs to Decision Tree classifier with the result of %67.21. ORCID NO: 0000-0002-4678-6972

Anahtar Kelimeler: Machine Learning, Heart Disease Prediction, Classification Algorithms



 


Keywords: