Research Article

A Software Engineering Framework for Intelligent Code Smell Detection with Machine Learning

Volume: 2 Number: 1 May 31, 2026
EN TR

A Software Engineering Framework for Intelligent Code Smell Detection with Machine Learning

Abstract

Identifying code smells is a fundamental task in software engineering for improving software quality and ensuring long-term system sustainability. This paper presents an intelligent framework designed to detect Large Class type code smells using a machine learning (ML) model trained on source code metrics. Within this proposed framework, multiple software metrics, including structural metrics, Halstead metrics, and cognitive metrics, are derived from Java source code files to serve as input features for classification. The Extreme Gradient Boosting (XGBoost) algorithm is utilized to identify whether a software contains a code smell or not. Experimental evaluations conducted on a dataset demonstrated that the proposed model reached an accuracy of 93.47%. Furthermore, it achieved an average improvement of 6.33% compared to existing methods evaluated on the same dataset. These findings indicate that our ML-based approach can provide effective and reliable solutions for automating code smell detection in software engineering.

Keywords

References

  1. M.G. Al-Obeidallah and D. Al-Fraihat. “A systematic review on software code smells,” International Journal of Electrical and Computer Engineering, vol. 15, no. 3, pp. 3010-3027, June 2025. doi: 10.11591/ijece.v15i3
  2. M. Hadj-Kacem and N. Bouassida, “Application of Deep Learning for Code Smell Detection: Challenges and Opportunities,” SN Computer Sciences, no 5, 614, 2024. doi: 10.1007/s42979-024-02956-5
  3. F.R. Santos and R. Choren. “Data preprocessing for machine learning based code smell detection: A systematic literature review,” Information and Software Technology, vol. 184, 2025. doi: 10.1016/j.infsof.2025.107752
  4. P. S. Thakur, S. S. Chouhan, S. S. Rathore and J. Parmar. “Systematic literature review on software code smell detection approaches,” Journal of Systems and Software, vol. 235, 2026, 112784. doi: 10.1016/j.jss.2026.112784
  5. S. Alodibat and M. Azzeh, “A Systematic Review on Code Smell Detection Approaches in Open Source Projects,” Software: Practice and Experience, vol. 1, pp. 1-30, 2026. doi: 10.1002/spe.70074
  6. P.S. Yadav, R.S. Rao, A. Mishra, and M. Gupta. “Machine Learning-Based Methods for Code Smell Detection: A Survey,” Applied Sciences, vol. 14, no. 14, 6149, 2024. doi: 10.3390/app14146149
  7. R. Gupta and S. K. Singh, "Investigating transfer learning for code smell detection on homogeneous data," Innovations in Systems and Software Engineering, vol. 22, no. 3, pp. 1-20, Jan. 2026. doi:10.1007/s11334-025-00624-2
  8. C. Kumar and S. K. Singh, "CodeSmellSynergy: A Hybrid Ensemble Framework for Python Code Smell Detection," Discover Computing, vol. 29, no. 171, pp. 1-25, 2026. doi:10.1007/s10791-026-10055-7

Details

Primary Language

English

Subjects

Artificial Intelligence (Other)

Journal Section

Research Article

Publication Date

May 31, 2026

Submission Date

May 9, 2026

Acceptance Date

May 21, 2026

Published in Issue

Year 2026 Volume: 2 Number: 1

APA
Birant, K. U. (2026). A Software Engineering Framework for Intelligent Code Smell Detection with Machine Learning. Innovative Artificial Intelligence, 2(1), 42-48. https://izlik.org/JA62AD22FR
AMA
1.Birant KU. A Software Engineering Framework for Intelligent Code Smell Detection with Machine Learning. INNAI. 2026;2(1):42-48. https://izlik.org/JA62AD22FR
Chicago
Birant, Kökten Ulaş. 2026. “A Software Engineering Framework for Intelligent Code Smell Detection With Machine Learning”. Innovative Artificial Intelligence 2 (1): 42-48. https://izlik.org/JA62AD22FR.
EndNote
Birant KU (May 1, 2026) A Software Engineering Framework for Intelligent Code Smell Detection with Machine Learning. Innovative Artificial Intelligence 2 1 42–48.
IEEE
[1]K. U. Birant, “A Software Engineering Framework for Intelligent Code Smell Detection with Machine Learning”, INNAI, vol. 2, no. 1, pp. 42–48, May 2026, [Online]. Available: https://izlik.org/JA62AD22FR
ISNAD
Birant, Kökten Ulaş. “A Software Engineering Framework for Intelligent Code Smell Detection With Machine Learning”. Innovative Artificial Intelligence 2/1 (May 1, 2026): 42-48. https://izlik.org/JA62AD22FR.
JAMA
1.Birant KU. A Software Engineering Framework for Intelligent Code Smell Detection with Machine Learning. INNAI. 2026;2:42–48.
MLA
Birant, Kökten Ulaş. “A Software Engineering Framework for Intelligent Code Smell Detection With Machine Learning”. Innovative Artificial Intelligence, vol. 2, no. 1, May 2026, pp. 42-48, https://izlik.org/JA62AD22FR.
Vancouver
1.Kökten Ulaş Birant. A Software Engineering Framework for Intelligent Code Smell Detection with Machine Learning. INNAI [Internet]. 2026 May 1;2(1):42-8. Available from: https://izlik.org/JA62AD22FR