Computatio : Journal of Computer Science and Information Systems
Vol. 10 No. 1 (2026): Computatio: Journal of Computer Science and Information Systems

Perbandingan Algoritma K-Nearest Neighbours (KNN), Decision Tree, dan Artifical Neural Network (ANN) dalam Klasifikasi Risiko Kesehatan Maternal

Carin Gunawan (a:1:{s:5:"en_US"
s:24:"Universitas Tarumanagara"
})



Article Info

Publish Date
16 Jul 2026

Abstract

Maternal health is a crucial phase, both for the mother's condition and for the prospective baby. In 2021, the maternal mortality rate has been on the rise. This is caused by various factors that can affect the physical health of pregnant women. These factors include age,, blood sugar, blood pressure, heart rate, and body temperature. Analysis and diagnosis of the still-unfavorable maternal physical condition trigger poor maternal health risks due to the lack of proper preventive care for the mother. Maternal health risks (y) are divided into three classes: low risk, medium risk, and high risk, which are the target classes (y). Grouping of maternal health risks is done by classification using three algorithms: K-Nearest Neighbors (KNN), Artificial Neural Network (ANN), and Decision Tree . The classification is carried out using a dataset from Kaggle, with 7 features (x) and 3 classes (y), namely 'low risk', 'mid risk', and 'high risk'. The research shows that the decision tree algorithm shows the highest accuracy value of 0.89, while the accuracy values for the decision tree and ANN algorithms are 0.71 and 0.72, respectively.

Copyrights © 2026