Fauziyah, Nadiyah Jihan, Rahmania, Fadilla, Daniyal, Muhammad and Sari, Nur Fitriyah Ayu Tunjung Sari (2024) Analisis dan optimalisasi performa algoritma gaussian naive bayes pada prediksi metabolic syndrome menggunakan SMOTE. JISKA(Jurnal Informatika Sunan Kalijaga), 9 (2). pp. 112-122. ISSN 2528 –0074
Text
20732.pdf Download (625kB) |
Abstract
Metabolic syndrome is a complex global health problem, with symptoms such as abdominal obesity, insulin resistance, high blood pressure, high blood sugar, and abnormal blood lipids. With this global challenge, several studies have attempted to predict these diseases using machine learning methods. However, often, predictions about a disease result in data imbalance where minority classes are underrepresented. To balance the class proportions, the Synthetic Minority Over-sampling Technique (SMOTE) method replicates the minority class samples. In this research, the technique applied to predict is the Gaussian Naive Bayes (GNB) algorithm. The results show an increase in prediction accuracy by 0.2 from 0.81 to 0.83. This study confirms the critical role of the SMOTE oversampling method in machine learning using the Gaussian Naive Bayes (GNB) algorithm in Metabolic Syndrome prediction and its positive impact on diagnostic efficiency and public health.
Item Type: | Journal Article |
---|---|
Keywords: | Metabolic Syndrome; Machine Learning; Gaussian Naive Bayes; Synthetic Minority Over-sampling Technique (SMOTE); Prediction |
Subjects: | 08 INFORMATION AND COMPUTING SCIENCES > 0801 Artificial Intelligence and Image Processing > 080109 Pattern Recognition and Data Mining 08 INFORMATION AND COMPUTING SCIENCES > 0801 Artificial Intelligence and Image Processing > 080110 Simulation and Modelling 08 INFORMATION AND COMPUTING SCIENCES > 0801 Artificial Intelligence and Image Processing |
Divisions: | Faculty of Technology > Department of Informatics Engineering |
Depositing User: | Nur Fitriyah Ayu Tunjung Sari |
Date Deposited: | 30 Oct 2024 09:44 |
Downloads
Downloads per month over past year
Origin of downloads
Actions (login required)
View Item |