KLASIFIKASI PENYAKIT MATA MENGGUNAKAN ENSEMBLE WEIGHTED AVERAGE PADA ARSITEKTUR INCEPTIONV3 MOBILENETV2 DAN XCEPTION

Naufaldi, Irsyad Rafi (2025) KLASIFIKASI PENYAKIT MATA MENGGUNAKAN ENSEMBLE WEIGHTED AVERAGE PADA ARSITEKTUR INCEPTIONV3 MOBILENETV2 DAN XCEPTION. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (Cover) COVER.pdf Download (861kB)
	Text (Bab 1) BAB 1.pdf Download (328kB)
	Text (Bab 2) BAB 2.pdf Restricted to Repository staff only until 4 December 2027. Download (1MB)
	Text (Bab 3) BAB 3.pdf Restricted to Repository staff only until 4 December 2027. Download (1MB)
	Text (Bab 4) BAB 4.pdf Restricted to Repository staff only until 4 December 2027. Download (1MB)
	Text (Bab 5) BAB 5.pdf Download (244kB)
	Text (Daftar Pustaka) DAFTAR PUSTAKA.pdf Download (191kB)
	Text (Lampiran) LAMPIRAN.pdf Restricted to Repository staff only Download (258kB)

Abstract

Eye health plays a crucial role in human life. Visual impairment can significantly reduce productivity and may even lead to blindness. In Indonesia, the limited number and uneven distribution of ophthalmologists often result in delayed diagnosis of eye diseases. Therefore, an AI-based system is needed to support early detection. This study aims to develop an eye disease classification system using deep learning with the Ensemble Weighted Average (EWA) method, which combines three top-performing Transfer Learning architectures: InceptionV3, MobileNetV2, and Xception. A fundus image dataset from Kaggle was used, consisting of four classes: cataract, diabetic retinopathy, glaucoma, and normal. The research process includes preprocessing (data splitting, resizing, and normalization), model training, and ensemble prediction using weights derived from each model’s accuracy. Evaluation was carried out using accuracy, precision, recall, and F1-score across two data split schemes (60:20:20 and 70:15:15). The results show that the Ensemble Weighted Average method improves accuracy to 92.22% using the 70:15:15 split, outperforming individual models. The system was then deployed into a Flask-based website to facilitate fast, accurate, and efficient early diagnosis of eye diseases.

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Rahajoe, Rr. Ani Dijah	NIDN0012057301	anidijah.if@upnjatim.ac.id
Thesis advisor	Puspaningrum, Eva Yulia	NIDN0005078908	evapuspaningrum.if@upnjatim.ac.id

Subjects:

R Medicine > R Medicine (General)
T Technology > T Technology (General)

Divisions:

Faculty of Computer Science > Departemen of Informatics

Depositing User:

Irsyad Rafi Naufaldi

Date Deposited:

04 Dec 2025 04:17

Last Modified:

04 Dec 2025 04:39

URI:

https://repository.upnjatim.ac.id/id/eprint/47778