Main Article Content

Abstract

The heart is one of the most important organs for human; therefore, it needs to always be looked after and maintained properly. If it is not looked after and maintained properly, it will be at risk of disease. Currently, heart disease of various types still ranks first in deaths both in Indonesia and abroad. Various efforts continue to be developed by relevant scientists to detect it. Considering the importance of development efforts, in this research a machine-learning program was designing to classify heart disease as a detection system effort. In this article we will describe the analysis of the characteristics of the K-NN classifier, decision tree, random forest (accuracy, precision and recall), as well as determining the best classifier for detecting heart disease. To support the analysis of test results, Python and Google Colab programming has been implementation here. The best results obtained from the analysis of the application of these three models are the Decision Tree Classifier with accuracy, precision and recall values ​​of 90%, 87% and 88% respectively. These results indicate that this model has been successfully developing.

Keywords

Machine learning, K-NN algorithm, decision tree, random forest

Article Details

How to Cite
1.
Hartono A, Dewi LA, Yuniarti E, Salsabila Tahta Hirani Putri, Harahap TS. Machine Learning Classification for Detecting Heart Disease with K-NN Algorithm, Decision Tree and Random Forest. EKSAKTA [Internet]. 2023Dec.30 [cited 2025Apr.3];24(04):513-22. Available from: https://eksakta.ppj.unp.ac.id/index.php/eksakta/article/view/461

References

Read More