PREDIKSI RISIKO DIABETES MENGGUNAKAN ENSEMBLE WEIGHTED VOTING DENGAN ALGORITMA LOGISTIC REGRESSION, SVM, DAN RANDOM FOREST

Prasetyo, Ach.Diki (2025) PREDIKSI RISIKO DIABETES MENGGUNAKAN ENSEMBLE WEIGHTED VOTING DENGAN ALGORITMA LOGISTIC REGRESSION, SVM, DAN RANDOM FOREST. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text 21081010055 COVER.pdf Download (855kB)
	Text (BAB 1) 21081010055 BAB 1.pdf Download (210kB)
	Text (BAB 2) 21081010055 BAB 2.pdf Restricted to Repository staff only until 19 June 2027. Download (626kB)
	Text (BAB 3) 21081010055 BAB 3.pdf Restricted to Repository staff only until 19 June 2027. Download (1MB)
	Text (BAB 4) 21081010055 BAB 4.pdf Restricted to Repository staff only until 19 June 2027. Download (1MB)
	Text (BAB 5) 21081010055 BAB 5.pdf Download (177kB)
	Text (DAFTAR PUSTAKA) 21081010055 DAFTAR PUSTAKA.pdf Download (168kB)

Abstract

Diabetes mellitus is a chronic disease that continues to increase in number of sufferers globally, so a preventive solution is needed such as the use of an early prediction system for diabetes risk. This study aims to develop and test a diabetes risk prediction model using the Ensemble Weighted Voting method that combines three machine learning algorithms, namely Logistic Regression, Support Vector Machine, and Random Forest. The data used comes from the 2021 "Behavioral Risk Factor Surveillance System" (BRFSS) survey taken from Kaggle, as well as primary data taken in Indonesia using Google Form. Model testing was carried out using a confusion matrix with evaluation metrics of accuracy, precision, recall, and F1-score. The test results show that the Ensemble model is able to provide better accuracy results compared to a single model. In scenarios 1 and 2, the ensemble model gets the highest accuracy with values of 74,89% and 74,57%. Then, in scenario 4, which is only trained using 5% of Kaggle data and tested using local data, the model gets the highest accuracy of 90,00%. Although in scenario 3 the random forest model obtained higher accuracy, overall the model using the ensemble showed better performance, with higher accuracy values than the single model in most scenarios. The diabetes prediction risk system has also been successfully implemented in the form of a website using Flask as the backend and Laravel as the frontend.

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Anggraeny, Fetty Tri	NIDN0711028201	fettyangraeny.if@upnjatim.ac.id
Thesis advisor	Mumpuni, Retno	NIDN0016078703	retnomumpuni.if@upnjatim.ac.id

Subjects:

Q Science > QA Mathematics > QA75 Electronic computers. Computer science

Divisions:

Faculty of Computer Science > Departemen of Informatics

Depositing User:

Ach.Diki Prasetyo

Date Deposited:

20 Jun 2025 02:39

Last Modified:

20 Jun 2025 02:39

URI:

https://repository.upnjatim.ac.id/id/eprint/38671