KOMPARASI INTEGRASI TEKNIK RESAMPLING PADA NAÏVE BAYES (NB) DAN LOGISTIC REGRESSION (LR) BERBASIS PARTICLE SWARM OPTIMIZATION UNTUK KLASIFIKASI CACAT PERANGKAT LUNAK

HARDONI, ANDRE and Rini, Dian Palupi and Sukemi, Sukemi (2021) KOMPARASI INTEGRASI TEKNIK RESAMPLING PADA NAÏVE BAYES (NB) DAN LOGISTIC REGRESSION (LR) BERBASIS PARTICLE SWARM OPTIMIZATION UNTUK KLASIFIKASI CACAT PERANGKAT LUNAK. Master thesis, Sriwijaya University.

	Text RAMA_55101_09042611822003.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (6MB) | Request a copy
Preview	Text RAMA_55101_09042611822003_0023027804_0003126604_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (1MB) | Preview
	Text RAMA_55101_09042611822003_0023027804_0003126604_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_55101_09042611822003_0023027804_0003126604_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_55101_09042611822003_0023027804_0003126604_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_55101_09042611822003_0023027804_0003126604_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (22kB) | Request a copy
	Text RAMA_55101_09042611822003_0023027804_0003126604_06.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (368kB) | Request a copy
	Text RAMA_55101_09042611822003_0023027804_0003126604_07_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (987kB) | Request a copy
	Text RAMA_55101_09042611822003_Turnitin.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (12MB) | Request a copy

Abstract

Software quality is generally used as a condition that must exist in a software developer or builder company to make the company more competitive. Software quality in general can be seen from the number of defects contained in the resulting software. Improving software quality can be done in various ways, but the best approach is to prevent defects, one of which is done by predicting the possibility of defects through the classification method. The NASA MDP dataset is public and is widely used by researchers for defect classification cases, but the lack of this dataset is class imbalance and noise attribute. This experiment introduces a combination method between SMOTE (Synthetic Minority Over-sampling Technique) and PSO (Particle Swarm Optimization) to deal with imbalance and noise attribute problems which are integrated into the basic method like Naive Bayes classification method and logistic regression. The results of experiments that have been carried out on 9 (nine) NASA MDP datasets show that overall SMOTE + PSO integration can improve classification performance with the highest average AUC (Area Under Curve) value in the logistic regression method, which is 0.853 and in the nave Bayes method, with 0.831 and its also better than without combining the two.

Item Type:	Thesis (Master)
Uncontrolled Keywords:	Cacat Perangkat Lunak, Klasifikasi, Naïve Bayes, Logistic Regression, SMOTE, PSO
Subjects:	Q Science > QA Mathematics > QA75-76.95 Calculating machines > QA76.9.B45 Big data. Machine learning. Quantitative research. Metaheuristics.
Divisions:	09-Faculty of Computer Science > 55101-Informatics (S2)
Depositing User:	ANDRE HARDONI
Date Deposited:	11 Jan 2022 07:34
Last Modified:	11 Jan 2022 07:34
URI:	http://repository.unsri.ac.id/id/eprint/61103

Actions (login required)

View Item