A Bedbug Optimization-Based Machine Learning Framework for Software Fault Prediction

Bahman Arasteh (), Seyed Salar Sefati, Eduard-Cristian Popovici (), Ibrahim Furkan Ince and Farzad Kiani
Additional contact information
Bahman Arasteh: Department of Software Engineering, Faculty of Engineering and Natural Science, Istinye University, Istanbul 34396, Türkiye
Seyed Salar Sefati: Department of Software Engineering, Faculty of Engineering and Natural Science, Istinye University, Istanbul 34396, Türkiye
Eduard-Cristian Popovici: Telecommunications Department, Faculty of Electronics, Telecommunications and Information Technology, National University of Science and Technology POLITEHNICA Bucharest, 060042 Bucharest, Romania
Ibrahim Furkan Ince: Department of Software Development, Faculty of Arts and Sciences, Beykent University, Istanbul 34500, Türkiye
Farzad Kiani: Data Science Application and Research Center (VEBIM), Fatih Sultan Mehmet Vakif University, Istanbul 34445, Türkiye

Mathematics, 2025, vol. 13, issue 21, 1-26

Abstract: Predicting software faults and identifying defective modules is a significant challenge in developing reliable software products. Machine Learning (ML) approaches on the historical fault datasets are utilized to classify faulty software modules. The presence of irrelevant features within the training datasets undermines the accuracy and precision of the software prediction models. Consequently, selecting the most effective features for module classification constitutes an NP-hard problem. This research introduces the Binary Bedbug Optimization Algorithm (BBOA) to extract the most effective features of training datasets. The primary contribution lies in the development of a binary variant of the Bedbug Optimization Algorithm (BOA) designed to effectively select effective features and build a classifier for identifying faulty software modules using ANN, SVM, DT, and NB algorithms. The model’s performance was evaluated using five standard real-world NASA datasets. The findings reveal that among the 21 features analyzed, features such as code complexity, lines of code, the total number of operands and operators, lines containing both code and comments, the total count of operators and operands, and the number of branch instructions play a critical role in predicting software faults. The proposed method achieved notable improvements, with increases of 5.97% in accuracy, 3.86% in precision, 2.37% in sensitivity (recall), and 3.06% in F1-score.

Keywords: fault prediction; binary bedbug optimization algorithm; feature selection; machine learning (search for similar items in EconPapers)
JEL-codes: C (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/2227-7390/13/21/3531/pdf (application/pdf)
https://www.mdpi.com/2227-7390/13/21/3531/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jmathe:v:13:y:2025:i:21:p:3531-:d:1787011

Access Statistics for this article

Mathematics is currently edited by Ms. Emma He

More articles in Mathematics from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().