Classification of Priority Recipients for RTLH Assistance Using LightGBM with Bayesian Optimization in Jombang Regency

Pratama, Ferdi Firdaus Ega (2026) Classification of Priority Recipients for RTLH Assistance Using LightGBM with Bayesian Optimization in Jombang Regency. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (Cover) 22081010247.-cover.pdf Download (1MB)
	Text (Bab 1) 22081010247.-bab1.pdf Download (223kB)
	Text (Bab 2) 22081010247.-bab2.pdf Restricted to Repository staff only until 12 June 2028. Download (354kB)
	Text (Bab 3) 22081010247.-bab3.pdf Restricted to Repository staff only until 12 June 2028. Download (893kB)
	Text (Bab 4) 22081010247.-bab4.pdf Restricted to Repository staff only until 12 June 2028. Download (1MB)
	Text (Bab 5) 22081010247.-bab5.pdf Download (145kB)
	Text (Daftar Pustaka) 22081010247.-daftarpustaka.pdf Download (172kB)
	Text (Lampiran) 22081010247.-lampiran.pdf Restricted to Repository staff only Download (209kB)

Abstract

The Uninhabitable Housing Assistance (RTLH) assistance program in Jombang Regency still faces challenges in determining aid recipient priorities because the selection process is conducted manually and involves many assessment variables. This study aims to develop a classification model for RTLH assistance recipient priorities using the Light Gradient Boosting Machine (LightGBM) algorithm with hyperparameter optimization through Bayesian Optimization. The dataset used in this study was obtained from the Department of Housing and Settlement Areas of Jombang Regency, consisting of 9,173 data records. After data cleaning, the dataset was reduced to 7,280 records with 36 variables representing housing physical conditions and household socio-economic characteristics. The preprocessing stage included missing value removal, categorical data transformation using Label Encoding, and numerical data normalization using Min-Max Scaler. The evaluation results showed that the baseline model achieved an accuracy of 93.44%, precision of 94.89%, recall of 93.95%, and F1-Score of 94.42%. After optimization using Bayesian Optimization, the model performance improved with an accuracy of 94.71%, precision of 95.26%, recall of 95.81%, and F1-Score of 95.53%. The optimized model was implemented into a Streamlit-based web application to support the initial selection process of RTLH assistance recipients more quickly and objectively. Overall, the combination of LightGBM and Bayesian Optimization improved the classification performance in determining RTLH assistance recipient priorities in Jombang Regency.

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Puspaningrum, Eva Yulia	NIDN0005078908	evapuspaningrum.if@upnjatim.ac.id
Thesis advisor	Al Haromainy, Muhammad Muharrom	NIDN0701069503	muhammad.muharrom.if@upnjatim.ac.id

Subjects:

Q Science > QA Mathematics > QA76.6 Computer Programming

Divisions:

Faculty of Computer Science > Departemen of Informatics

Depositing User:

Ferdi Firdaus Ega Pratama

Date Deposited:

15 Jun 2026 03:26

Last Modified:

15 Jun 2026 03:26

URI:

https://repository.upnjatim.ac.id/id/eprint/53915