Permadani, Citra Amelia Intan (2025) Penerapan Metode HDBSCAN (Hierarchical Density-Based Spatial Clustering of Application with Noise) Untuk Pengelompokan Provinsi di Indonesia Berdasarkan Indikator Ketenagakerjaan dalam Analisis Daya Saing Angkatan Kerja. Undergraduate thesis, UPN Veteran Jawa Timur.
![[img]](https://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (Cover)
21083010004-COVER.pdf Download (983kB) |
|
Text (Bab 1)
21083010004-BAB 1.pdf Download (189kB) |
|
Text (Bab 2)
21083010004-BAB 2.pdf Restricted to Repository staff only until 18 September 2027. Download (344kB) |
|
Text (Bab 3)
21083010004-BAB 3.pdf Restricted to Repository staff only until 18 September 2027. Download (443kB) |
|
Text (Bab 4)
21083010004-BAB 4.pdf Restricted to Repository staff only until 18 September 2027. Download (1MB) |
|
Text (Bab 5)
21083010004-BAB 5.pdf Download (181kB) |
|
Text (Daftar Pustaka)
21083010004-DAFTAR PUSTAKA.pdf Download (163kB) |
|
Text (Lampiran)
21083010004-LAMPIRAN.pdf Restricted to Repository staff only until 18 September 2027. Download (469kB) |
Abstract
The labor force consists of people aged 15 years and above who are active in economic activities and reflects the potential of human resources who are ready to work. However, Indonesia faces challenges in the form of an increase in the labor force that is not proportional to the availability of jobs and low educational qualifications that hamper competitiveness, as stated in Presidential Regulation No. 12 of 2025. The productive age variable (15–64 years) has a higher participation rate, better education correlates with the formal sector and more decent wages, while job availability affects labor mobility. This study aims to apply the HDBSCAN (Hierarchical Density-Based Spatial Clustering of Applications with Noise) method to group provinces in Indonesia based on employment indicators, including productive age, education graduates, and employment sectors, to analyze the competitiveness of the provincial workforce. The HDBSCAN method was chosen because it is capable of exploring various density scales to form stable clusters and identify and separate outliers commonly found in labor data due to differences in social and economic conditions between provinces. With the best parameters of min_cluster_size = 2 and the Euclidean metric, a Silhouette Coefficient value of 0.546 was obtained with three main clusters and one noise cluster, namely cluster 0 consisting of 3 provinces, cluster 1 consisting of 25 provinces, cluster 2 consisting of 3 provinces, and cluster -1 (noise) consisting of 3 provinces. These results confirm that the distribution of employment in Indonesia is uneven, especially in terms of education levels and the dominance of employment sectors. This study also developed an interactive web-based GUI using the Streamlit framework to visualize the clustering results in the form of interactive maps.
| Item Type: | Thesis (Undergraduate) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||||||
| Subjects: | Q Science > QA Mathematics > QA76.6 Computer Programming | ||||||||||||
| Divisions: | Faculty of Computer Science > Departemen of Data Science | ||||||||||||
| Depositing User: | Citra Amelia Intan Permadani | ||||||||||||
| Date Deposited: | 19 Sep 2025 03:28 | ||||||||||||
| Last Modified: | 19 Sep 2025 03:28 | ||||||||||||
| URI: | https://repository.upnjatim.ac.id/id/eprint/43799 |
Actions (login required)
 |
View Item |