Fauzizah, Deanita Nur (2026) Implementasi Hibrida K-Prototypes Dengan Ant Colony Optimization (ACO) untuk Pengelompokan Prioritas Penanganan Stunting Berdasarkan Kabupaten/Kota di Indonesia. Undergraduate thesis, UPN Veteran Jawa Timur.
![[img]](https://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (Cover)
22083010005-cover.pdf Download (1MB) |
|
Text (Bab 1)
22083010005-bab1.pdf Download (212kB) |
|
Text (Bab 2)
22083010005-bab2.pdf Restricted to Repository staff only until 16 September 2028. Download (499kB) |
|
Text (Bab 3)
22083010005-bab3.pdf Restricted to Repository staff only until 16 September 2028. Download (573kB) |
|
Text (Bab 4)
22083010005-bab4.pdf Restricted to Repository staff only until 16 September 2028. Download (2MB) |
|
Text (Bab 5)
22083010005-bab5.pdf Download (165kB) |
|
Text (Daftar Pustaka)
22083010005-daftarpustaka.pdf Download (146kB) |
|
Text (Lampiran)
22083010005-lampiran.pdf Restricted to Repository staff only until 16 September 2028. Download (509kB) |
Abstract
Stunting is a health issue that requires tailored interventions based on the characteristics of each region. Differences in conditions across regions mean that the needs for addressing stunting cannot be treated uniformly across all districts and cities. This study aims to classify districts and cities in Indonesia based on indicators related to stunting and to determine priorities for addressing the issue using the K-Prototypes method with Ant Colony Optimization (ACO). The data used comprised 514 districts/cities with categorical variables stunting prevalence and wasting prevalence and numerical variables underweight prevalence, access to safe drinking water, access to adequate sanitation, and completion of basic immunizations. The K-Prototypes method was used because it can process both numerical and categorical data, while ACO was used to optimize the gamma parameter that influences the distance calculation process in K-Prototypes. Determining the number of clusters using the Elbow Method resulted in three clusters. The ACO optimization yielded an optimal gamma value of 14,78. Evaluation using the Silhouette Score and the Davies-Bouldin Index (DBI) showed that K-Prototypes with ACO optimization yielded a Silhouette Score of 0.3691 and a DBI of 1.1535, which were better than those of K-Prototypes without optimization, which yielded a Silhouette Score of 0.3286 and a DBI of 1.2063. The clustering results yielded three clusters: Cluster 1, a high-priority area comprising 116 regencies/cities; Cluster 2, an area in good condition comprising 177 regencies/cities; and Cluster 3, a medium-priority area comprising 221 regencies/cities. These results indicate that applying ACO optimization to K-Prototypes can improve the quality of clustering and provide an overview of regional characteristics that can serve as a basis for determining priorities in addressing stunting.
| Item Type: | Thesis (Undergraduate) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||||||
| Subjects: | Q Science > Q Science (General) Q Science > QA Mathematics Q Science > QA Mathematics > QA75 Electronic computers. Computer science |
||||||||||||
| Divisions: | Faculty of Computer Science > Departemen of Data Science | ||||||||||||
| Depositing User: | Deanita Nur Fauzizah | ||||||||||||
| Date Deposited: | 16 Sep 2026 06:33 | ||||||||||||
| Last Modified: | 16 Sep 2026 08:35 | ||||||||||||
| URI: | https://repository.upnjatim.ac.id/id/eprint/60279 |
Actions (login required)
 |
View Item |