OPTIMASI K-MEDOIDS DENGAN PARTICLE SWARM OPTIMIZATION DALAM PENENTUAN PRIORITAS BANTUAN PEMERINTAH DI DESA KALIPURO

Fitriani, Aulia Nur (2025) OPTIMASI K-MEDOIDS DENGAN PARTICLE SWARM OPTIMIZATION DALAM PENENTUAN PRIORITAS BANTUAN PEMERINTAH DI DESA KALIPURO. Undergraduate thesis, Universitas Pembangunan Nasional Veteran Jawa Timur.

	Text (Cover) 21083010051_Cover.pdf Download (1MB)
	Text (Bab I) 21083010051_Bab I.pdf Download (55kB)
	Text (Bab II) 21083010051_Bab II.pdf Restricted to Repository staff only until 17 March 2028. Download (379kB)
	Text (Bab III) 21083010051_Bab III.pdf Restricted to Repository staff only until 17 March 2028. Download (189kB)
	Text (Bab IV) 21083010051_Bab IV.pdf Restricted to Repository staff only until 17 March 2028. Download (1MB)
	Text (Bab V) 21083010051_Bab V.pdf Download (14kB)
	Text (Daftar Pustaka) 21083010051_Daftar Pustaka.pdf Download (132kB)
	Text (Lampiran) 21083010051_Lampiran.pdf Restricted to Repository staff only until 17 March 2028. Download (32kB)

Abstract

The distribution of government assistance in Indonesia is often hampered by inaccuracies in recipient data between those recorded in government systems and field conditions. In Kalipuro Village, Mojokerto District, data mismatches caused difficulties in screening assistance, requiring village officials to manually re-filter the data. This triggered protests from residents who should have received assistance but did not get their rights. To overcome this problem, this research proposes the use of the K-Medoids algorithm which is able to overcome sensitivity to outliers. This algorithm is used to cluster data based on criteria such as the number of vehicle assets, property, and income. In addition, this research incorporates the Particle Swarm Optimization (PSO) algorithm to optimize the clustering process, which is expected to improve accuracy and efficiency in social assistance distribution. Based on the evaluation results, the Silhouette Score value increases as the number of clusters increases, with K = 5 producing the highest score of 0.8199. This shows that the formation of five clusters provides a more optimal structure than the other number of clusters. The clustering results show that cluster 1 with 89 families is prioritized as the main beneficiary, followed by cluster 2 with 94 families, cluster 4 with 296 families, cluster 3 with 356 families, and cluster 0 with 177 families. The total number of members of the five clusters is 1012 families. Each cluster represents a group of beneficiaries with similar characteristics, obtained through the application of the Particle Swarm Optimization (PSO) and K-Medoids Clustering algorithms.

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Hindrayani, Kartika Maulida	NIDN0009099205	kartikamaulida.ds@upnjatim.ac.id
Thesis advisor	Trimono, Trimono	NIDN0005128001	trimono.stat@upnjatim.ac.id

Subjects:

Q Science > QA Mathematics
Q Science > QA Mathematics > QA76.6 Computer Programming

Divisions:

Faculty of Computer Science > Departemen of Data Science

Depositing User:

Aulia Nur Fitriani

Date Deposited:

17 Mar 2025 04:50

Last Modified:

17 Mar 2025 04:50

URI:

https://repository.upnjatim.ac.id/id/eprint/35574