Deteksi Kecacatan Perangkat Lunak Menggunakan Support Vector Machine Teroptimasi Berbasis Grey Wolf Optimizer dan Random Walk

Siswantoro, Muhammad Zain Fawwaz Nuruddin (2025) Deteksi Kecacatan Perangkat Lunak Menggunakan Support Vector Machine Teroptimasi Berbasis Grey Wolf Optimizer dan Random Walk. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 6025222009-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2027. Download (10MB) | Request a copy

Abstract

Deteksi kecacatan perangkat lunak merupakan proses penting dalam pengembangan perangkat lunak untuk mengidentifikasi sebagai bug sehingga aplikasi dapat berfungsi tanpa kesalahan. Namun, proses ini sering memakan biaya dan waktu. Penelitian ini mengusulkan penggunaan Support Vector Machine (SVM) yang hyperparameter-nya dioptimasi menggunakan Grey Wolf Optimizer (GWO) yang dipadukan dengan Random Walk (RW). Selain itu penelitian ini juga menggunakan Principal Component Analysis (PCA) sebagai pengurangan dimensi fitur dan juga oversampling dengan Synthetic Minority Over-sampling Technique (SMOTE) untuk menyeimbangkan dataset. Hasil dari penelitian ini menunjukan bahwa GWO yang dipadukan dengan RW mampu meningkatkan akurasi SVM dalam mengklasifikasi deteksi kecacatan perangkat lunak dibandingkan dengan optimasi lain, dengan akurasi berkisar antara 76,26% - 98,21% dan rata-rata akurasi 87,03% pada berbagai dataset, sehingga membuktikan efektivitasnya dalam deteksi kecacatan perangkat lunak.
=================================================================================================================================
Software defect detection is an important process in software development to identify as a bug so that the application can function without errors. However, this process is often costly and time consuming. This study proposes the use of Support Vector Machine (SVM) whose hyperparameters are optimized using Grey Wolf Optimizer (GWO) combined with Random Walk (RW). In addition, this study also uses Principal Component Analysis (PCA) as a feature dimension reduction and also oversampling with Synthetic Minority Over-sampling Technique (SMOTE) to balance the dataset. The results of this study show that GWO combined with RW is able to increase the accuracy of SVM in classifying software defect detection compared to other optimizations, with an accuracy ranging from 76.26% - 98.21% and an average accuracy of 87.03% on various datasets, thus proving its effectiveness in software defect detection.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	Support Vector Machine, Deteksi Kecacatan Perangkat Lunak, Grey Wolf Optimization, Optimasi Hyperparameter, Principal Component Analysis, Random Walk ====================================================== Support Vector Machine, Software Defect Detection, Grey Wolf Optimization, Hyperparameter Optimization, Principal Component Analysis, Random Walk
Subjects:	T Technology > T Technology (General) > T57.8 Nonlinear programming. Support vector machine. Wavelets. Hidden Markov models. T Technology > T Technology (General) > T58.62 Decision support systems
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55101-(S2) Master Thesis
Depositing User:	Muhammad Zain Fawwaz Nuruddin Siswantoro
Date Deposited:	06 Feb 2025 04:24
Last Modified:	06 Feb 2025 04:24
URI:	http://repository.its.ac.id/id/eprint/118354

Actions (login required)

View Item