Penerapan Smote-Enn dan XGBoost dengan Optimasi Bayesian Untuk Mengatasi Imbalance Class Dalam Deteksi Penyakit Gagal Jantung

Pramesti, Bintang Tiara (2025) Penerapan Smote-Enn dan XGBoost dengan Optimasi Bayesian Untuk Mengatasi Imbalance Class Dalam Deteksi Penyakit Gagal Jantung. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (Cover) 21081010333.-cover.pdf Download (1MB)
	Text (Bab 1) 21081010333.-bab1.pdf Download (266kB)
	Text (Bab 2) 21081010333.-bab2.pdf Restricted to Repository staff only Download (624kB)
	Text (Bab 3) 21081010333.-bab3.pdf Restricted to Repository staff only until 15 September 2027. Download (836kB)
	Text (Bab 4) 21081010333.-bab4.pdf Restricted to Repository staff only until 15 September 2027. Download (892kB)
	Text (Bab 5) 21081010333.-bab5.pdf Download (256kB)
	Text (Daftar Pustaka) 21081010333.-daftar pustaka.pdf Download (184kB)
	Text (Lampiran) 21081010333.-lampiran.pdf Restricted to Repository staff only Download (266kB)

Abstract

Class imbalance is a common issue in applying machine learning to medical data, especially in detecting heart failure. This condition makes models biased toward the majority class (healthy) while ignoring the minority (heart failure). This study addresses the problem by applying SMOTE-ENN and comparing the performance of XGBoost under three scenarios: without balancing, with Default SMOTE-ENN, and with Fine-Tuned SMOTE-ENN using Bayesian Optimization (BO). The dataset was preprocessed, outliers handled, split into 70:15:15, and evaluated using Accuracy, Precision, Recall, F1-score, and Stratified K-Fold Cross-Validation. The results show that XGBoost without SMOTE achieved high Accuracy (94.27%) but had poor Recall (09.17%). Applying Default SMOTE-ENN significantly improved performance with Recall of 94.08% and F1-score of 92.91%, while Fine-Tuned SMOTE-ENN BO achieved the highest Recall (94.24%) with an F1-score of 94.71%. Although Accuracy was slightly lower than the default, the fine-tuned model proved more effective in detecting minority cases. In conclusion, combining SMOTE-ENN with Bayesian Optimization enhances the sensitivity of XGBoost and provides an effective strategy to address class imbalance in medical datasets

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Anggraeny, Fetty Tri	NIDN0711028201	fettyanggraeny.if@upnjatim.ac.id
Thesis advisor	Saputra, Wahyu Syaifullah Jauharis	NIDN0725088601	wahyu.s.j.saputra.if@upnjatim.ac.id

Subjects:

Q Science > QA Mathematics > QA76.6 Computer Programming

Divisions:

Faculty of Computer Science > Departemen of Informatics

Depositing User:

Bintang Tiara Pramesti

Date Deposited:

15 Sep 2025 06:36

Last Modified:

15 Sep 2025 06:36

URI:

https://repository.upnjatim.ac.id/id/eprint/43522