KLASIFIKASI PENYAKIT KULIT MENGGUNAKAN ALGORITMA ENSEMBEL RF-DCNN(RANDOM FOREST-DEEP CONVOLUTIONAL NEURAL NETWORK

Kurniawan, Ananda Rheza (2025) KLASIFIKASI PENYAKIT KULIT MENGGUNAKAN ALGORITMA ENSEMBEL RF-DCNN(RANDOM FOREST-DEEP CONVOLUTIONAL NEURAL NETWORK. Undergraduate thesis, Universitas Pembangunan Nasional Veteran Jawa Timur.

	Text (Cover) 21081010141_Cover.pdf Download (2MB)
	Text (Bab 1) 21081010141_Bab I.pdf Download (134kB)
	Text (Bab 2) 21081010141_Bab II.pdf Restricted to Registered users only until 12 June 2028. Download (884kB) | Request a copy
	Text (Bab 3) 21081010141_Bab III.pdf Restricted to Registered users only until 12 June 2028. Download (609kB) | Request a copy
	Text (Bab 4) 21081010141_Bab IV.pdf Restricted to Registered users only until 12 June 2028. Download (1MB) | Request a copy
	Text (Bab 5) 21081010141_Bab V.pdf Restricted to Registered users only until 12 June 2028. Download (35kB) | Request a copy
	Text (Daftar Pustaka) 21081010141_Daftar Pustaka.pdf Download (148kB)

Abstract

Skin diseases such as psoriasis, tinea, and atopic dermatitis often present with visually similar symptoms, making them difficult to distinguish through manual observation. This study aims to develop an image-based skin disease classification system using Deep learning and ensemble learning approaches to improve diagnostic accuracy. The dataset comprises 4,235 skin images categorized into four classes: normal, psoriasis, tinea, and atopic dermatitis. The research implements a Convolutional Neural Network (CNN) model and an ensemble technique by combining Deep CNN with the Random Forest algorithm (RF-DCNN). To enhance model performance, Hyperparameter Tuning is applied using Bayesian Optimization to determine optimal configurations such as the number of filters, learning rate, and tree depth. Evaluation is conducted using two data split schemes (80:10:10 and 70:20:10) for training, validation, and testing phases. Experimental results indicate that the RF-DCNN method achieved the highest performance, with accuracy rates of 84% and 83% in the respective data split scenarios. These findings demonstrate that the integration of Deep learning, ensemble methods, and Hyperparameter Tuning effectively improves the classification of skin disease images and holds potential for application in clinical decision support systems.

Item Type:	Thesis (Undergraduate)
Contributors:	Contribution Contributors NIDN/NIDK Email Thesis advisor Via, Yisti Vita NIDN0025048602 yistivia.if@upnjatim.ac.id UNSPECIFIED Nurlaili, Afina Lina NIDN0013129303 afina.lina.if@upnjatim.com
Subjects:	T Technology > T Technology (General)
Divisions:	Faculty of Computer Science > Departemen of Informatics
Depositing User:	Ananda Rheza Kurniawan
Date Deposited:	12 Jun 2025 08:04
Last Modified:	12 Jun 2025 08:04
URI:	https://repository.upnjatim.ac.id/id/eprint/37416

Actions (login required)

View Item