PENERAPAN TEKNIK RANDOM OVERSAMPLING UNTUK MENGATASI IMBALANCED CLASS DATA PADA KLASIFIKASI TINGKAT KEBUGARAN TUBUH MANUSIA MENGGUNAKAN METODE K-NEAREST NEIGHBOR

SAPUTRA, RACHMAN DIMAS and Palupi Rini, Dian and Primanita, Anggina (2023) PENERAPAN TEKNIK RANDOM OVERSAMPLING UNTUK MENGATASI IMBALANCED CLASS DATA PADA KLASIFIKASI TINGKAT KEBUGARAN TUBUH MANUSIA MENGGUNAKAN METODE K-NEAREST NEIGHBOR. Undergraduate thesis, Sriwijaya University.

	Text RAMA_55201_09021281823055.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (5MB) | Request a copy
	Text RAMA_55201_09021281823055_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (8MB) | Request a copy
Preview	Text RAMA_55201_09021281823055_0023027804_0206088901_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (2MB) | Preview
	Text RAMA_55201_09021281823055_0023027804_0206088901_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (276kB) | Request a copy
	Text RAMA_55201_09021281823055_0023027804_0206088901_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (336kB) | Request a copy
	Text RAMA_55201_09021281823055_0023027804_0206088901_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (406kB) | Request a copy
	Text RAMA_55201_09021281823055_0023027804_0206088901_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_55201_09021281823055_0023027804_0206088901_06.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (88kB) | Request a copy
	Text RAMA_55201_09021281823055_0023027804_0206088901_07_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (225kB) | Request a copy
	Text RAMA_55201_09021281823055_0023027804_0206088901_08_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (456kB) | Request a copy

Abstract

Physical fitness is the body's ability to carry out physical activities without causing excessive fatigue, knowing a level of body fitness is useful for determining the right solution to overcome problems related to body fitness. A classification process with machine learning is needed to determine the level of body fitness, one method that is often used is the K-NN method. Classification using machine learning often has a problem with class imbalance that causes errors in classification, to overcome this problem a data balancing method is needed. The random oversampling technique is a data balancing technique by randomly adding minority class samples until the number of samples is equal to the majority class. The results obtained after applying random oversampling decreased the average accuracy by 4% and the average precision by 6%, but there was an increase in the average recall value by 4%. This research proves that the random oversampling technique can make the recall value higher.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Klasifikasi, Kebugaran Tubuh, Data Tidak Seimbang, Random Oversampling, K-Nearest Neighbor
Subjects:	Q Science > QA Mathematics > QA75-76.95 Calculating machines > QA76.76.I58.A3115 Computer science. Computers. Intelligent agents (Computer software)
Divisions:	09-Faculty of Computer Science > 55201-Informatics (S1)
Depositing User:	mr Rachman Dimas Saputra
Date Deposited:	25 Jan 2023 05:56
Last Modified:	25 Jan 2023 05:56
URI:	http://repository.unsri.ac.id/id/eprint/87448

Actions (login required)

View Item