Munir, Salma Fathiyatur Rizky (2026) Performance Analysis of the KNN and SVM Algorithms in Stunting Classification Based on Anthropometric Standards. Undergraduate thesis, UPN Veteran Jawa Timur.
|
Text (Cover)
Cover.pdf Download (1MB) | Preview |
|
|
Text (Chapter I)
Chapter I.pdf Download (120kB) | Preview |
|
![[img]](https://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (Chapter II)
Chapter II.pdf Restricted to Repository staff only until 25 May 2028. Download (281kB) |
|
|
Text (Chapter III)
Chapter III.pdf Restricted to Repository staff only until 25 May 2028. Download (394kB) |
|
|
Text (Chapter IV)
Chapter IV.pdf Restricted to Repository staff only until 25 May 2028. Download (655kB) |
|
|
Text (Chapter V)
Chapter V.pdf Download (112kB) | Preview |
|
|
Text (References)
References.pdf Download (164kB) | Preview |
|
|
Text (Appendix)
Appendix.pdf Restricted to Repository staff only Download (86kB) |
Abstract
The purpose of this study is to evaluate how well the KNN and SVM algorithms in toddlers at the Patianrowo Community Health Center. Since stunting is a long-term nutritional issue that can impair children’s growth and development, categorization techniques are essential for accurately identifying stunting cases. This study compares the two algorithms performance using evaluation metrics and examines how parameter changes affect classification performance. The Patianrowo Community Health Center provided primary data on newborns between the ages of 0 and 60 months. After preprocessing, 1,067 of the original 1,102 data points were recovered these were then split into 20% test data and 80% training data. Gender, age, height, weight, and Z-score values are among the features that are employed. While the SVM algorithm employed linear, polynomial, and radial basis function (RBF) kernels, the KNN algorithm testing was tested with several K values, specifically 3, 5, 7, 9, and 11. A confusion matrix comprising assessment metrics including accuracy, precision, recall, and F1-score was used to asse the model’s performance. According to the test findings, KNN with K = 5 generated an F1-score of 77.52%, recall of 67.73%, accuracy of 96.72%, and precision of 91.25. With an accuracy of 97.47%, precision of 90.82%, recall of 78.96%, and an F1-score of 82.55%, the polynomial kernel produced the best result for the SVM algorithm. These findings indicate that the SVM technique with a polynomial kernel outperforms the others in stunting classification.
| Item Type: | Thesis (Undergraduate) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||||||
| Subjects: | T Technology > T Technology (General) T Technology > T Technology (General) > T58.6-58.62 Management Information Systems |
||||||||||||
| Divisions: | Faculty of Computer Science > Departemen of Informatics | ||||||||||||
| Depositing User: | Salma Fathiyatur Rizky Munir | ||||||||||||
| Date Deposited: | 26 May 2026 04:08 | ||||||||||||
| Last Modified: | 26 May 2026 05:39 | ||||||||||||
| URI: | https://repository.upnjatim.ac.id/id/eprint/52638 |
Actions (login required)
 |
View Item |