Comparing Effectiveness Of Static Vs. Change Metrics To Analyze The Impact On Model Performance In Predicting Fault Prone Software Modules
DOI:
https://doi.org/10.29284/zsq5qe53Keywords:
Software Fault Prediction, Hybrid Learning, Change Metrics, Genetic Algorithms, Deep Learning.Abstract
Predicting fault‑prone modules is a key challenge in software engineering. Early identification of risky components helps teams allocate testing resources effectively, reduce costs, and improve reliability. Two major categories of predictors are static metrics — measures derived from source code without version history — and change (evolution) metrics — measures based on historical evolution from version control systems. This paper investigates and compares the predictive effectiveness of static vs. change metrics in identifying fault‑prone modules. We review related work, define theoretical foundations, describe a comparative empirical methodology, and discuss outcomes, limitations, and future research directions. Results indicate that while static metrics offer useful baseline insight, change metrics often provide superior predictive power for fault proneness due to their historical context.
References
1. Basili, V.R., Briand, L.C., and Melo, W.L. (1996). A Validation of Object Oriented Design Metrics as Quality Indicators. IEEE Transactions on Software Engineering.
2. Hassan, A.E. (2009). Predicting Faults Using the Complexity of Code Changes. International Conference on Software Engineering.
3. Kim, S., Zimmermann, T., and Nagappan, N. (2007). An Empirical Study of Code Churn and Faults. ESEC/FSE.
4. McCabe, T.J. (1976). A Complexity Measure. IEEE Transactions on Software Engineering.
5. Menzies, T., et al. (2010). Defect Prediction from Static Code Features: Current State and Future Direction. IEEE Transactions on Software Engineering.
6. Nagappan, N., Ball, T., & Zeller, A. (2005). Mining Metrics to Predict Component Failures. International Conference on Software Engineering. Zimmermann, T., et al. (2009). Cross-project defect prediction: a large scale experiment on data vs. domain vs. process. ESEC/FSE.
7. Lessmann, S., et al. (2008). Benchmarking classification models for software defect prediction: A proposed framework and novel findings. IEEE TSE.
8. Alenezi, M., & Zarour, M. (2024). Semantic and traditional feature fusion for software defect prediction using hybrid deep learning model. Scientific Reports.
9. Koya, J. R., et al. (2024). Software Defect Prediction Using Hybrid Machine Learning Techniques: A Comparative Study. SN Computer Science.
10. Nam, J., et al. (2013). Heterogeneous defect prediction. IEEE Transactions on Software Engineering.
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
This work is licensed under a Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
