Qasim Ali¹, Asma Zubedi², Fatima Najmuddin¹*, Imran Memon3 and Salahuddin Sadar¹

¹Mehran University of Engineering and Technology, Jamshoro, Software Department, Pakistan
²Beijing University of Post and Telecommunication, Management and Beijing, China
³Bahria University Karachi Campus, Computer Science Department, Karachi, Pakistan

*Corresponding Author: Fatima Najmuddin, Mehran University of Engineering and Technology, Jamshoro, Software Department, Pakistan.

Received: March 22, 2023; Published: March 29, 2023

Abstract

The dependency on software has been increasing with each passing day, due to which reliability and quality of software has been becoming more and more crucial. The quality of product increases when the defects and faults will decrease. To find the defect in software product, many approaches were proposed but machine learning approach is very useful. Machine learning classify data into defective and non-defective modules.
In this paper, 15 datasets from NASA promise repository named AR1, AR3, AR5, AR6, CM1, KC1, KC2, KC3, MC1, MC2, MW1, PC1, PC2, PC3, PC4 are analyzed using 7 machine learning algorithms (K-Nearest Neighbor, Linear regression, Random Forest, Naïve Bayes, Support Vector Machine (SVM), logistic regression and decision tree) and apply 10 k-fold cross validation in RapidMiner tool. In RapidMiner performance of ML algorithm in term of accuracy is calculated and the summary of the result shows that SVM perform best.
Similarly, these 15 datasets are also analyzed using 8 machine learning algorithms named Random Forest, Naïve Bayes, simple logistic, Sequential minimal optimization (SMO), K-star, REF (decision tree), K-Nearest Neighbor (KNN), decision table and apply 10 k-fold cross validation in WEKA simulation tool. In WEKA performance of ML algorithm in term of correctly classified instances is calculated and the summary of the result shows that Random Forest and simple logistic perform best.

Keywords: Software Defect; Machine Learning Algorithms; Weka; RapidMiner; Nasa Promise Datasets

References

1. Challagulla B., et al. "Empirical assessment of machine learning based software defect prediction techniques”. International Journal on Artificial Intelligence Tools 17 (2008): 389-400.
2. Z Zhang., et al. "Multiple kernel ensemble learning for software defect prediction”. Automatic Software Engineering 23 (2016): 569-590.
3. P Deep singh and A Chug. "Software defect prediction anlysis using machine learning algorithm". in 2017 7^th International Conference on Cloud Computing, Data Science and Engineering-Confluence, IEEE (2017): 775-781.
4. N Kalaivani and D R Beena. "Overview of software defect prediction using machine learning algorithms”. International Journal of Pure and Applied Mathematics 118 (2018): 3863-3871.
5. A Hammouri., et al. "Software bug prediction using machine learning approach”. International Journal of Advanced Computer Science and Applications 9 (2018): 78-83.
6. Y Soe., et al. "Software Defect Prediction Using Random Forest Algorithm”. 2018 12^th South East Asian Technical University Consortium (SEATUC) 1 (2018): 1-5.
7. A Alsaeedi and M Zubair Khan. "Software defect prediction using supervised machine learning and ensemble techniques: A comparative study”. Journal of Software Engineering and Applications 12 (2019): 85-100.
8. C Lakshmi Prabha and D N Shivakumar "Software defect prediction using machine learning techniques". in 2020 4^th International Conference on Trends in Electronics and Informatics (ICOEI) (48184), IEEE (2020): 728-733.
9. M Shah and N Pujara. "A Review On Software Defects Prediction Methods". arXiv (2020): 1-5.

Citation

Citation: Fatima Najmuddin., et al. “Comparative Study: To Analyze Software Defects Using Machine Learning Algorithms". Acta Scientific Computer Sciences 5.4 (2023): 103-108.

Copyright

Copyright: © 2023 Fatima Najmuddin., et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.