Implementasi Metode Affinity Propagation (AP) dengan Principal Component Analysis (PCA) dalam Clustering Data Pasien Hipertensi

Bhalqis, Anissa Andiar (2026) Implementasi Metode Affinity Propagation (AP) dengan Principal Component Analysis (PCA) dalam Clustering Data Pasien Hipertensi. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (Cover) 21083010038_Cover.pdf Download (765kB)
	Text (Bab 1) 21083010038_Bab1.pdf Download (95kB)
	Text (Bab 2) 21083010038_Bab2.pdf Restricted to Repository staff only until 9 March 2029. Download (237kB)
	Text (Bab 3) 21083010038_Bab3.pdf Restricted to Repository staff only until 9 March 2029. Download (250kB)
	Text (Bab 4) 21083010038_Bab4.pdf Restricted to Repository staff only until 9 March 2029. Download (1MB)
	Text (Bab 5) 21083010038_Bab5.pdf Download (67kB)
	Text (Daftar Pustaka) 21083010038_Daftar Pustaka.pdf Download (93kB)
	Text (Lampiran) 21083010038_Lampiran.pdf Restricted to Repository staff only until 9 March 2029. Download (74kB)

Abstract

Hypertension is a chronic disease characterized by an increase in blood pressure in the arteries beyond normal levels. Data from the Surabaya Health Profile for 2019–2021 shows an increase in the prevalence of hypertension from 70.9% to 84.9%, confirming that this condition is a serious health problem because it reduces quality of life and increases the burden on health services. Data from the Indonesian Ministry of Health (2024) shows that the cost burden on BPJS due to hypertension reached IDR 22.8 trillion with a low screening rate (3 out of 10 people), while the WHO (2023) shows that only 15% of patients receive treatment and 4% have their blood pressure controlled. This condition indicates that hypertension management is not yet optimal and remains uniform, potentially leading to interventions that are not targeted and a waste of health resources. Therefore, to support a more efficient treatment approach, a data-driven approach is needed to group patients based on risk factors. Affinity Propagation is an automatic clustering algorithm that determines cluster centers without setting the initial number of clusters. The purpose of this study is to group patients with hypertension. AP works through the exchange of responsibility and availability messages between data to determine cluster centers based on similarity. To improve the quality of grouping, PCA is applied as a dimension reduction stage to reduce data redundancy and noise, namely variables with small variations that are often considered less important. The reduction results show that the four initial variables are summarized into two main components. With AP parameters of preference −500, damping 0.9, and maximum iterations of 200, seven clusters were obtained with a Silhouette Coefficient of 0.39, indicating moderate clustering quality. Thus, the combination of PCA and Affinity Propagation approaches has the potential to support more targeted medical decision-making in the management of hypertension.

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Hindrayani, Kartika Maulida	NIDN0009099205	kartika.maulida.ds@upnjatim.ac.id
Thesis advisor	Idhom, Mohammad	NIDN0010038305	idhom@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:

Anissa Andiar Bhalqis

Date Deposited:

10 Mar 2026 01:28

Last Modified:

10 Mar 2026 01:28

URI:

https://repository.upnjatim.ac.id/id/eprint/50284