Klasterisasi Tingkat Kemiskinan Menggunakan Fuzzy C-Means Dengan Optimalisasi Entropy weight Dan PCA

Hidayat, Jihan Octavia Salsabillah (2025) Klasterisasi Tingkat Kemiskinan Menggunakan Fuzzy C-Means Dengan Optimalisasi Entropy weight Dan PCA. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (Cover) 21081010034-cover.pdf Download (889kB)
	Text (BAB I) 21081010034-bab1.pdf Download (99kB)
	Text (BAB II) 21081010034-bab2.pdf Restricted to Repository staff only until 28 May 2027. Download (242kB)
	Text (BAB III) 21081010034-bab3.pdf Restricted to Repository staff only until 28 May 2027. Download (343kB)
	Text (BAB IV) 21081010034-bab4.pdf Restricted to Repository staff only until 28 May 2027. Download (5MB)
	Text (BAB V) 21081010034-bab5.pdf Download (79kB)
	Text (Daftar pustaka) 21081010034-daftarpustaka.pdf Download (27kB)

Abstract

Poverty is one of the major issues faced by East Java Province, ranking as the third-highest in Indonesia in terms of poverty rate. To understand the patterns and characteristics of poverty in this region, an analytical method capable of accurately clustering areas based on socioeconomic indicators is required. This study aims to cluster the poverty levels in districts/cities in East Java using the Fuzzy C-Means (FCM), optimized with Entropy weight and Principal Component Analysis (PCA). The analysis process begins with feature selection using the entropy weight method, which determines the objective weight of each poverty indicator based on its variation. Out of 14 initial variables, six key variables were selected: Number of Poor Population, Percentage of Poor Population, P1 (Poverty Depth Index), P2 (Poverty Severity Index), Poverty Line, and Per Capita Expenditure. Subsequently, dimensionality reduction using PCA was performed, resulting in two main components (PC1 and PC2) that explain 86.01% of the total information in the dataset. The clustering stage using Fuzzy C-Means (FCM) determines the optimal number of clusters based on the Silhouette score, with results showing three clusters and a Silhouette score of 0.55. Cluster 1 consists of 10 districts/cities with the lowest poverty rate, Cluster 2 includes 21 districts/cities with a middle-income economy, and Cluster 3 consists of 7 districts/cities with the highest poverty rate. The model evaluation using Silhouette score resulted in a score of 0.55, indicating that the clustering results are adequate and appropriate. Cluster 1 has a more compact membership distribution, while clusters 2 and 3 show a broader socioeconomic spread.

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Rahajoe, Ani Dijah	NIDN0012057301	anidijah.if@upnjatim.ac.id
Thesis advisor	Haromainy, Muhammad Muharrom Al	NIDN0701069503	muhammad.muharrom.if@upnjatim.ac.id

Subjects:

J Political Science > JS Local government Municipal government
Q Science > Q Science (General)
Q Science > QA Mathematics > QA76 Computer software
Q Science > QA Mathematics > QA76.6 Computer Programming

Divisions:

Faculty of Computer Science > Departemen of Informatics

Depositing User:

Jihan Octavia

Date Deposited:

28 May 2025 07:45

Last Modified:

28 May 2025 07:45

URI:

https://repository.upnjatim.ac.id/id/eprint/36916