KLASIFIKASI GAMBAR PADA GOOGLE QUICK DRAW MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORK DAN NEUROEVOLUTION OF AUGMENTING TOPOLOGIES

NURJAMIL, NURJAMIL and Primartha, Rifkie and Miraswan, Kanda Januar (2019) KLASIFIKASI GAMBAR PADA GOOGLE QUICK DRAW MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORK DAN NEUROEVOLUTION OF AUGMENTING TOPOLOGIES. Undergraduate thesis, Sriwijaya University.

	Text RAMA_55201_09021281419056.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (9MB) | Request a copy
Preview	Text RAMA_55201_09021281419056_0001067709_0009019002_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (2MB) | Preview
	Text RAMA_55201_09021281419056_0001067709_0009019002_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_55201_09021281419056_0001067709_0009019002_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_55201_09021281419056_0001067709_0009019002_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (2MB) | Request a copy
	Text RAMA_55201_09021281419056_0001067709_0009019002_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_55201_09021281419056_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (16MB) | Request a copy
	Text RAMA_55201_09021281419056_0001067709_0009019002_07_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (881kB) | Request a copy
	Text RAMA_55201_09021281419056_0001067709_0009019002_08_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (4MB) | Request a copy
	Text RAMA_55201_09021281419056_0001067709_0009019002_06.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy

Abstract

Klasifikasi gambar Google quick draw bukanlah suatu tugas yang mudah, dikarenakan banyaknya varian dan noise pada data. Pada penelitian ini, kami membuat Pengklasifikasi sketsa gambar dari basis data Google Quick Draw dan membandingkan performa antara metode Convolutional Neural Network(CNN) dan Neuroevolution of Augmenting Topologies(NEAT). Penelitian membuktikan bahwa teknik konvolusi yang terdapat pada metode CNN terbukti lebih unggul dibandingkan dengan teknik genetik algoritma atau algoritma evolusi yang digunakan pada metode NEAT. Hasil menunjukkan CNN mendapatkan performa yang lebih baik dengan rata – rata akurasi yang didapatkan sebesar 89.125% sedangkan metode NEAT mendapatkan rata – rata akurasi sebesar 85.8125%. Meskipun demikian satu dari delapan kelas yang diprediksi menggunakan metode NEAT dapat lebih unggul dari metode CNN sebesar 4.5%.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Klasifikasi gambar, Convolutional Neural Network, Neural Network, Neuroevolution, Neuroevolution of Augmenting Topologies
Subjects:	Q Science > Q Science (General) > Q300-390 Cybernetics > Q325.5 Machine learning
Divisions:	09-Faculty of Computer Science > 55201-Informatics (S1)
Depositing User:	Users 4460 not found.
Date Deposited:	17 Jan 2020 07:46
Last Modified:	17 Jan 2020 07:46
URI:	http://repository.unsri.ac.id/id/eprint/24430

Actions (login required)

View Item