FEBRIANTO, ISHAK (2023) Komparasi Algoritma Decision Tree C4.5 Dan Naive Bayes Pada Klasifikasi Status Gizi Balita Stunting. Undergraduate thesis, UNIVERSITAS PEMBANGUNAN NASIONAL “VETERAN” JAWA TIMUR.
|
Text (Cover)
19081010072-cover.pdf Download (546kB) | Preview |
|
|
Text (Bab 1)
19081010072-bab1.pdf Download (190kB) | Preview |
|
![[img]](http://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (Bab 2)
19081010072-bab2.pdf Restricted to Repository staff only until 31 May 2025. Download (460kB) |
|
|
Text (Bab 3)
19081010072-bab3.pdf Restricted to Repository staff only until 31 May 2025. Download (530kB) |
|
|
Text (Bab 4)
19081010072-bab4.pdf Restricted to Repository staff only until 31 May 2025. Download (1MB) |
|
|
Text (Bab 5)
19081010072-bab5.pdf Download (170kB) | Preview |
|
|
Text (Daftar pustaka)
19081010072-daftarpustaka.pdf Download (199kB) | Preview |
|
|
Text (Lampiran)
19081010072-lampiran.pdf Restricted to Repository staff only until 31 May 2025. Download (716kB) |
Abstract
Nutritional problems remain a crucial issue today. One of the many nutritional problems is stunting. Stunting, according to WHO (2015), is a condition of malnutrition or inadequate nutrition during a child's growth and development, characterized by a height or length below the standard. Data mining with classification techniques on the nutritional status of stunted toddlers can be done to help identify toddlers who experience stunting and provide objective measurements of their nutritional status. In this study, the C4.5 decision tree algorithm, which falls under the decision tree approach, and naive Bayes, which uses a probability-based approach to class occurrence, will be compared in classifying the nutritional status of stunted toddlers. Discretization will be performed during the preprocessing stage. The data used in this study were obtained from Jagir Community Health Center, Surabaya, and consisted of secondary data on the nutritional status of toddlers in 2021, totaling 2801. Stunting or normal labeling in the dataset used the standard anthropometric references for children in Indonesia as stated in the Indonesian Ministry of Health Regulation number 2 of 2020. If the z-score of a child's anthropometric measurement is less than minus 2 standard deviations (SD) based on the length-for-age (LAZ) or height-for-age (HAZ) index, the child is classified as stunted. The comparison of models was analyzed based on three scenarios of training and testing data division (70:30, 80:20, 90:10). The classification models were evaluated using accuracy, precision, recall, f1-score, and AUC score in the ROC curve for each scenario. The results showed that in the best scenario with a 70:30 training and testing data ratio, the decision tree C4.5 outperformed naive Bayes based on an AUC score of 86% (good classification), while naive Bayes produced an AUC score of 77% (fair classification) in classifying the nutritional status of stunted toddlers.
| Item Type: | Thesis (Undergraduate) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||||||
| Subjects: | T Technology > T Technology (General) | ||||||||||||
| Divisions: | Faculty of Computer Science > Departemen of Informatics | ||||||||||||
| Depositing User: | Ishak Febrianto | ||||||||||||
| Date Deposited: | 06 Jun 2023 05:19 | ||||||||||||
| Last Modified: | 06 Jun 2023 05:19 | ||||||||||||
| URI: | http://repository.upnjatim.ac.id/id/eprint/14303 |
Actions (login required)
 |
View Item |