Pengelompokan Spatio-Temporal pada Kasus Tuberkulosis di Kota Surabaya Menggunakan Generalized LASSO

Sari, Putri Kartika (2025) Pengelompokan Spatio-Temporal pada Kasus Tuberkulosis di Kota Surabaya Menggunakan Generalized LASSO. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5003211102_Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (3MB) | Request a copy

Abstract

Tuberkulosis masih menjadi masalah kesehatan utama, dan Indonesia merupakan negara dengan jumlah kasus tertinggi kedua di dunia. Di Kota Surabaya, tingginya angka kasus dipengaruhi oleh kepadatan penduduk, mobilitas penduduk yang tinggi, serta kondisi sanitasi dan ventilasi rumah yang buruk. Beberapa penelitian yang telah dilakukan dalam pengelompokan kasus tuberkulosis di Indonesia, hanya memperoleh hasil cluster wilayah dan belum melibatkan efek spatio-temporal. Pada penelitian ini mengidentifikasi pola penyebaran tuberkulosis dengan mempertimbangkan aspek spasial dan temporal. Metode statistik yang digunakan adalah generalized LASSO karena metode ini dapat memastikan pemodelan yang akurat dalam mengidentifikasi pola penyebaran tuberkulosis di Kota Surabaya yang terhubung baik secara spasial maupun temporal. Variabel yang digunakan adalah jumlah kasus tuberkulosis di tiap kecamatan di Kota Surabaya pada bulan Januari 2022 hingga Desember 2024. Hasil penelitian menunjukkan bahwa pemilihan parameter optimal menggunakan Generalized Cross-Validation (GCV) mampu mengidentifikasi efek spasial dan temporal yang lebih baik dibandingkan Approximate Leave-One-Out Cross-Validation (ALOCV). Dengan nilai λ_T=2,20 dan λ_S=10,21 dan diperoleh tiga cluster untuk pengelompokan spasial dan empat cluster pada temporal.
========================================================================================================================
Tuberculosis continues to be a significant public health concern, with Indonesia ranking second globally in terms of the number of cases. In Surabaya, the high number of cases is influenced by population density, high population mobility, and poor sanitation and ventilation conditions in homes. A review of extant studies on tuberculosis case clustering in Indonesia reveals that prior research has exclusively identified spatial clusters and has not considered the effects of space and time. This study identifies tuberculosis spread patterns by considering spatial and temporal aspects. The statistical method employed is generalized LASSO, a technique that has been demonstrated to ensure accurate modeling in identifying tuberculosis spread patterns in Surabaya City, which are interconnected both spatially and temporally. The variables employed in this study encompass the number of tuberculosis cases in each subdistrict of Surabaya City from January 2022 to December 2024. The findings of the study suggest that the selection of optimal parameters using Generalized Cross-Validation (GCV) is more effective in identifying spatial and temporal effects compared to Approximate Leave-One-Out Cross-Validation (ALOCV). Utilizing the parameters of λ_T=2,20 and λ_S=10,21, the data was segmented into three clusters for spatial and four clusters for temporal.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Approximate Leave-One-Out Cross-Validation, Generalized Cross-Validation, Generalized LASSO, Pengelompokan Spatio-Temporal, Tuberkulosis, Approximate Leave-One-Out Cross-Validation, Generalized Cross-Validation, Generalized LASSO, Spatio-Temporal Clustering, Tuberculosis.
Subjects:	H Social Sciences > HA Statistics > HA30.6 Spatial analysis H Social Sciences > HA Statistics > HA31.7 Estimation R Medicine > RA Public aspects of medicine > RA0421 Public health. Hygiene. Preventive Medicine
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Putri Kartika Sari
Date Deposited:	04 Aug 2025 08:04
Last Modified:	04 Aug 2025 08:04
URI:	http://repository.its.ac.id/id/eprint/126957

Actions (login required)

View Item