Multi-view learning for software defect prediction

Background: Traditionally, machine learning algorithms have been simply applied for software defect prediction by considering single-view data, meaning the input data contains a single feature vector. Nevertheless, different software engineering data sources may include multiple and partially indep...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Elife Ozturk Kiyak, Derya Birant, Kokten Ulas Birant
Formato:	article
Lenguaje:	EN
Publicado:	Wroclaw University of Science and Technology 2021
Materias:	Software defect prediction multi-view learning machine learning k-nearest neighbors Computer software QA76.75-76.765
Acceso en línea:	https://doaj.org/article/24ab5bfe8ea24ec68f62a57a46c2184d
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:24ab5bfe8ea24ec68f62a57a46c2184d
record_format	dspace
spelling	oai:doaj.org-article:24ab5bfe8ea24ec68f62a57a46c2184d2021-11-05T15:05:33ZMulti-view learning for software defect prediction 10.37190/e-Inf2101082084-48401897-7979https://doaj.org/article/24ab5bfe8ea24ec68f62a57a46c2184d2021-08-01T00:00:00Zhttps://www.e-informatyka.pl/attach/e-Informatica_-_Volume_15/eInformatica2021Art08.pdfhttps://doaj.org/toc/1897-7979https://doaj.org/toc/2084-4840 Background: Traditionally, machine learning algorithms have been simply applied for software defect prediction by considering single-view data, meaning the input data contains a single feature vector. Nevertheless, different software engineering data sources may include multiple and partially independent information, which makes the standard single-view approaches ineffective. Objective: In order to overcome the single-view limitation in the current studies, this article proposes the usage of a multi-view learning method for software defect classification problems. Method: The Multi-View k-Nearest Neighbors (MVKNN) method was used in the software engineering field. In this method, first, base classifiers are constructed to learn from each view, and then classifiers are combined to create a robust multi-view model. Results: In the experimental studies, our algorithm (MVKNN) is compared with the standard k-nearest neighbors (KNN) algorithm on 50 datasets obtained from different software bug repositories. The experimental results demonstrate that the MVKNN method outperformed KNN on most of the datasets in terms of accuracy. The average accuracy values of MVKNN are 86.59%, 88.09%, and 83.10% for the NASA MDP, Softlab, and OSSP datasets, respectively. Conclusion: The results show that using multiple views (MVKNN) can usually improve classification accuracy compared to a single-view strategy (KNN) for software defect prediction. Elife Ozturk KiyakDerya BirantKokten Ulas Birant Wroclaw University of Science and TechnologyarticleSoftware defect predictionmulti-view learningmachine learningk-nearest neighborsComputer softwareQA76.75-76.765ENe-Informatica Software Engineering Journal, Vol 15, Iss 1 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Software defect prediction multi-view learning machine learning k-nearest neighbors Computer software QA76.75-76.765
spellingShingle	Software defect prediction multi-view learning machine learning k-nearest neighbors Computer software QA76.75-76.765 Elife Ozturk Kiyak Derya Birant Kokten Ulas Birant Multi-view learning for software defect prediction
description	Background: Traditionally, machine learning algorithms have been simply applied for software defect prediction by considering single-view data, meaning the input data contains a single feature vector. Nevertheless, different software engineering data sources may include multiple and partially independent information, which makes the standard single-view approaches ineffective. Objective: In order to overcome the single-view limitation in the current studies, this article proposes the usage of a multi-view learning method for software defect classification problems. Method: The Multi-View k-Nearest Neighbors (MVKNN) method was used in the software engineering field. In this method, first, base classifiers are constructed to learn from each view, and then classifiers are combined to create a robust multi-view model. Results: In the experimental studies, our algorithm (MVKNN) is compared with the standard k-nearest neighbors (KNN) algorithm on 50 datasets obtained from different software bug repositories. The experimental results demonstrate that the MVKNN method outperformed KNN on most of the datasets in terms of accuracy. The average accuracy values of MVKNN are 86.59%, 88.09%, and 83.10% for the NASA MDP, Softlab, and OSSP datasets, respectively. Conclusion: The results show that using multiple views (MVKNN) can usually improve classification accuracy compared to a single-view strategy (KNN) for software defect prediction.
format	article
author	Elife Ozturk Kiyak Derya Birant Kokten Ulas Birant
author_facet	Elife Ozturk Kiyak Derya Birant Kokten Ulas Birant
author_sort	Elife Ozturk Kiyak
title	Multi-view learning for software defect prediction
title_short	Multi-view learning for software defect prediction
title_full	Multi-view learning for software defect prediction
title_fullStr	Multi-view learning for software defect prediction
title_full_unstemmed	Multi-view learning for software defect prediction
title_sort	multi-view learning for software defect prediction
publisher	Wroclaw University of Science and Technology
publishDate	2021
url	https://doaj.org/article/24ab5bfe8ea24ec68f62a57a46c2184d
work_keys_str_mv	AT elifeozturkkiyak multiviewlearningforsoftwaredefectprediction AT deryabirant multiviewlearningforsoftwaredefectprediction AT koktenulasbirant multiviewlearningforsoftwaredefectprediction
_version_	1718444148512522240

Multi-view learning for software defect prediction

Ejemplares similares