PENANGANAN KETIDAKSEIMBANGAN DATA MENGGUNAKAN METODE SMOTE UNTUK MENINGKATKAN KEAKURATAN DALAM PREDIKSI HUBUNGAN BIDANG STUDI DENGAN PEKERJAAN PADA LULUSAN UNIVERSITAS SRIWIJAYA

KARTIKASARI, ANNISA and Desiani, Anita and Yahdin, Sugandi (2019) PENANGANAN KETIDAKSEIMBANGAN DATA MENGGUNAKAN METODE SMOTE UNTUK MENINGKATKAN KEAKURATAN DALAM PREDIKSI HUBUNGAN BIDANG STUDI DENGAN PEKERJAAN PADA LULUSAN UNIVERSITAS SRIWIJAYA. Undergraduate thesis, Sriwijaya University.

	Text RAMA_44201_08011281621045.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_44201_08011281621045_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (7MB) | Request a copy
Preview	Text RAMA_44201_08011281621045_0011127702_0027075803_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (940kB) | Preview
	Text RAMA_44201_08011281621045_0011127702_0027075803_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (377kB) | Request a copy
	Text RAMA_44201_08011281621045_0011127702_0027075803_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (217kB) | Request a copy
	Text RAMA_44201_08011281621045_0011127702_0027075803_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (898kB) | Request a copy
	Text RAMA_44201_08011281621045_0011127702_0027075803_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (215kB) | Request a copy
	Text RAMA_44201_08011281621045_0011127702_0027075803_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (443kB) | Request a copy

Abstract

The tracer study dataset of Universitas Sriwijaya is the result of a questionnaire completed by graduates of Universitas Sriwijaya from Career Development Center Unsri (CDC-Unsri). The tracer study dataset has 2934 data of graduates from academic years 2014-2016. There are 12 atributes in the data set such as a level of learning in lectures, a research project experience, an internship experience,a fieldwork experience, a level competence of education background, general education skill, an english skill, an internet user skill, a computer user skill, a research skill, a communication skill and a relevance of the education background with the graduates employment. The atribute label in this research is atribute a relevance of the education background with the graduates employmet but there is an imbalance data for this atribute because the distribution of data in some classes is very small compared to other classes. This atribute consist of 5 classes namely the very closely label has 214 data, the closely label has 304 data, the quite closely label has 475 data, the less closely label has 463 data , and the not closely label has 787 data. The amount of data in each class that is not balanced can affect the value of accuration in the prediction such as accuracy, precision and recall so it can be hard for find the information for each class. In this study the SMOTE method will be used to make synthetic data in the minority class so that the amount of data becomes balanced. Based on the test classification with KNN the value of accuracy, precision, and recall increased while in the ANN and SVM classification the accuracy value decreased but the value of precision and recall increased so it was concluded the SMOTE method did not always increase accuracy but could increase the value of precision and recall so that it could obtain information on each label.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Ketidakseimbangan Data, SMOTE, Keakuratan, Prediksi, Hubungan Bidang Studi dengan Pekerjaan
Subjects:	Q Science > QA Mathematics > QA75-76.95 Calculating machines > QA76.9.D343 Data mining. Database searching. Big data.
Divisions:	08-Faculty of Mathematics and Natural Science > 44201-Mathematics (S1)
Depositing User:	Users 7889 not found.
Date Deposited:	09 Sep 2020 06:27
Last Modified:	09 Sep 2020 06:27
URI:	http://repository.unsri.ac.id/id/eprint/34717

Actions (login required)

View Item