Comparison of Data Mining Algorithms Using Artificial Neural Networks (ANN) and Naive Bayes for Preterm Birth Prediction. (November 2020)

• Record Type: Journal Article
• Authors: Puspitasari, Diah; Ramanda, Kresna; Supriyatna, Adi; Wahyudi, Mochamad; Sikumbang, Erma Delima; Sukmana, Sulaeman Hadi
• Is Part Of: Journal of physics. Volume 1641(2020)
• Publication Date: 2020-11
• Subjects: Physics -- Congresses; 530.5
• DOI: 10.1088/1742-6596/1641/1/012068
• Languages: English
• ISSNs: 1742-6588

Abstract

Abstract: Premature birth is still a big problem in Indonesia, in general, 15 million babies are born prematurely every year, more than 1 million babies die from complications due to premature birth. The main purpose of this study is to compare the Artificial Neural Network and Naive Bayes datamining algorithm models to predict preterm birth so as to obtain clinical evidence in preterm birth long before confinement so that sudden preterm birth can be converted to normal nativity. The model proposed in research on the prediction of preterm birth is by applying an Artificial Neural Network (ANN) algorithm and Naive Bayes algorithm. Where the two algorithms will be compared the level of accuracy and the value of the AUC against the prediction of preterm birth The results obtained that the prediction of preterm birth using the Artificial Neural Network (ANN) algorithm produces an accuracy value of 90.67% and an ROC value of 0.954. While the Naive Bayes algorithm produces an accuracy value of 84.53% and an ROC value of 0.929. For this reason, it can be concluded that the Artificial Neural Network (ANN) algorithm has a superior accuracy of 6.14% and 0.025 for its ROC value in predicting preterm birth.