PEMILIHAN CLUSTER OPTIMAL UNTUK PENGELOMPOKKAN PROVINSI RAWAN KRIMINALITAS DI INDONESIA BERDASARKAN 5 METODE EVALUASI MENGGUNAKAN K-MEANS CLUSTERING

ANGGARA, RISKA TRI and Hanum, Herlina and Andriani, Yuli (2025) PEMILIHAN CLUSTER OPTIMAL UNTUK PENGELOMPOKKAN PROVINSI RAWAN KRIMINALITAS DI INDONESIA BERDASARKAN 5 METODE EVALUASI MENGGUNAKAN K-MEANS CLUSTERING. Undergraduate thesis, Sriwijaya University.

Preview	Image RAMA_44201_08011282126049_COVER.jpeg - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (222kB) | Preview
	Text RAMA_44201_08011282126049.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (2MB) | Request a copy
	Text RAMA_44201_08011282126049_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (6MB) | Request a copy
	Text RAMA_44201_08011282126049_0008016503_0002077202_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (1MB)
	Text RAMA_44201_08011282126049_0008016503_0002077202_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (4MB) | Request a copy
	Text RAMA_44201_08011282126049_0008016503_0002077202_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_44201_08011282126049_0008016503_0002077202_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_44201_08011282126049_0008016503_0002077202_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (558kB) | Request a copy
	Text RAMA_44201_08011282126049_0008016503_0002077202_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (424kB) | Request a copy
	Text RAMA_44201_08011282126049_0008016503_0002077202_07_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (328kB) | Request a copy

Abstract

Cluster analysis is one method that is usually used to group objects into groups that are similar to each other based on their characteristics. This study aims to cluster 34 provinces in Indonesia based on the number of crime cases in 2022 using the K-Means clustering method. Provinces are grouped based on crime cases, which include 8 variables, namely murder, domestic violence, rape, kidnapping, theft, narcotics, fraud, and public order violations. Determination of the optimal number of clusters with k=2, 3, 4, and 5 clusters was done by evaluation using five methods, namely Calinski Harabasz Index, Davies Bouldin Index, Elbow, Silhouette Coefficient, and Standard Deviation ratio. The results showed that the optimal number of clusters using the Calinski Harabasz Index, elbow, and standard deviation ratio methods was 4 clusters, while the Davies Bouldin Index and Silhouette Coefficient method was 2 clusters. Selection of the best clusters from 5 evaluation methods based on the results that have the most similarity because it is concluded that the results obtained are more representative and more accurate so that the best cluster selected is 4 clusters. Clustering in cluster 1 consists of 2 provinces consisting of North Sumatra and East Java Provinces with the variables of rape and fraud cases at a high level, while the other 6 variables are at a very high level. Cluster 2 consists of 3 provinces consisting of DKI Jakarta, West Java and South Sulawesi Provinces with the variables of rape, kidnapping and fraud cases at a very high level, while the other 5 variables are at a high level. Cluster 3 consists of 6 provinces consisting of Aceh, Riau, South Sumatra, Lampung, Central Java, and North Sulawesi with all variables at a high level. The remaining provinces are in cluster 4, which consists of 23 provinces with all variables at a medium level.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	K-Means Clustering, Kriminalitas, Evaluasi cluster
Subjects:	Q Science > QA Mathematics > QA1-939 Mathematics > QA37.3.1.64 Applied Mathematics
Divisions:	08-Faculty of Mathematics and Natural Science > 44201-Mathematics (S1)
Depositing User:	Riska Tri Anggara
Date Deposited:	20 Mar 2025 04:52
Last Modified:	20 Mar 2025 04:52
URI:	http://repository.unsri.ac.id/id/eprint/169472

Actions (login required)

View Item