KLASIFIKASI RAS MONGOLOID MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORK

HANDICO, DELFIN and Fachrurrozi, Muhammad and Primanita, Anggina (2023) KLASIFIKASI RAS MONGOLOID MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORK. Undergraduate thesis, Sriwijaya University.

	Text RAMA_55201_09021381823077.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (2MB) | Request a copy
	Text RAMA_55201_09021381823077_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (4MB) | Request a copy
	Text RAMA_55201_09021381823077_0222058001_0206088901_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (1MB)
	Text RAMA_55201_09021381823077_0222058001_0206088901_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (285kB) | Request a copy
	Text RAMA_55201_09021381823077_0222058001_0206088901_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (159kB) | Request a copy
	Text RAMA_55201_09021381823077_0222058001_0206088901_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (338kB) | Request a copy
	Text RAMA_55201_09021381823077_0222058001_0206088901_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (84kB) | Request a copy
	Text RAMA_55201_09021381823077_0222058001_0206088901_06.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (14kB) | Request a copy
	Text RAMA_55201_09021381823077_0222058001_0206088901_07_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (77kB) | Request a copy
	Text RAMA_55201_09021381823077_0222058001_0206088901_08_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (360kB) | Request a copy

Abstract

Mongoloid is a type of human race that exists in the world. Each mongoloid race has different characteristics and physical traits including eyes, nose, and skin color. This research aims to build a system that can classify mongoloid races from input images. This system uses the Convolutional Neural Network method which then the error results from the classification using the Confusion Matrix calculation. The dataset used is image data with a total of 1600 images divided into 1280 training data and 320 testing data. Experiments were carried out by changing the learning rate and epoch parameters of the model that had been built. Based on the experiments conducted in this study, the highest accuracy classification rate of 90% was obtained from a combination of a learning rate of 0.0001 and an epoch of 15 with an accuracy value of 90%, precision 95%, recall 87%, and f1-score 91%. So the combination of a lower learning rate value and a larger number of epochs provides a higher level of accuracy for classification.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Klasifikasi, Convolutional Neural Network, Ras Mongoloid, Confusion Matrix
Subjects:	T Technology > TR Photography > TR624-835 Applied photography Including artistic, commercial, medical photography, photocopying processes T Technology > TR Photography > TR925-1050 Photomechanical processes
Divisions:	09-Faculty of Computer Science > 55201-Informatics (S1)
Depositing User:	Delfin Handico
Date Deposited:	16 Jun 2023 04:59
Last Modified:	16 Jun 2023 04:59
URI:	http://repository.unsri.ac.id/id/eprint/110941

Actions (login required)

View Item