Improving Software Effort Estimation Through Feature Selection and Optimized SVR

Rahmi Putri; Gusti Eka Yuliastuti; Citra Nurina Prabiantissa

doi:10.30871/jaic.v10i4.13378

Authors

Rahmi Putri Institute Technology of Adhi Tama Surabaya
Gusti Eka Yuliastuti Institute Technology of Adhi Tama Surabaya
Citra Nurina Prabiantissa Institute Technology of Adhi Tama Surabaya

DOI:

https://doi.org/10.30871/jaic.v10i4.13378

Keywords:

Feature Selection, Machine Learning, Parameter Optimization, Software Effort Estimation, Support Vector Regression, Whale Optimization Algorithm

Abstract

Accurate software development effort estimation is essential but often hindered by high-dimensional data and the inefficiencies of handling feature selection and parameter tuning as separate, sequential processes. This study proposes an integrated Whale Optimization Algorithm–Support Vector Regression (WOA-SVR) framework that simultaneously optimizes binary feature selection and continuous SVR hyperparameters (C, γ, ϵ) within a unified search process. Evaluated using the NASA93 dataset under a strict nested 10-fold cross-validation protocol to prevent information leakage, the proposed model's performance was comprehensively assessed using six metrics (MMRE, MdMRE, Pred (25), RMSE, MAE, MAPE), accompanied by mean and standard deviation reporting. A rigorous ablation study empirically proved that simultaneous optimization outperforms sequential approaches. The proposed WOA-SVR successfully eliminated 7 redundant features, reducing the dimensionality from 22 to 15, and achieved an MMRE of 21.32% ± 3.25% and a Pred (25) of 64.52% ± 3.90%. It significantly outperformed Standard SVR, PSO-SVR, GA-SVR, and GWO-SVR. Statistical validation via the Wilcoxon Signed-Rank Test and Cliff’s Delta effect size confirmed large and practically significant improvements. While the results demonstrate that simultaneous optimization provides a robust and simplified alternative for early-stage estimation, the effectiveness is bounded to the NASA93 dataset, necessitating future validation on modern agile repositories.

Downloads

Download data is not yet available.

References

[1] M. Rahman, H. Sarwar, A. Kader, T. Goncalves, and T. T. Tin, “Review and Empirical Analysis of Machine Learning-Based Software Effort Estimation,” IEEE Access, vol. 12, pp. 85661–85680, 2024, doi: 10.1109/ACCESS.2024.3404879.

[2] M. Azzeh, A. B. Nassif, and I. B. Attili, “Predicting Software Effort from Use Case Points: A Systematic Review,” Sci. Comput. Program., Apr. 2021, doi: 10.1016/j.scico.2020.102596.

[3] A. Ali and C. Gravino, “Improving software effort estimation using bio-inspired algorithms to select relevant features: An empirical study,” Sci. Comput. Program., vol. 205, May 2021, doi: 10.1016/j.scico.2021.102621.

[4] C. H. Rashid et al., “Software Cost and Effort Estimation: Current Approaches and Future Trends,” IEEE Access, vol. 11, pp. 99268–99288, 2023, doi: 10.1109/ACCESS.2023.3312716.

[5] Y. Mahmood, N. Kama, A. Azmi, A. S. Khan, and M. Ali, “Software effort estimation accuracy prediction of machine learning techniques: A systematic performance evaluation,” Softw. Pract. Exp., vol. 52, no. 1, pp. 39–65, Jan. 2022, doi: 10.1002/spe.3009.

[6] J. Li, S. Sun, L. Xie, C. Zhu, and D. He, “Multi-kernel support vector regression with improved moth-flame optimization algorithm for software effort estimation,” Sci. Rep., vol. 14, no. 1, Dec. 2024, doi: 10.1038/s41598-024-67197-1.

[7] P. Phannachitta, “On an optimal analogy-based software effort estimation,” Inf. Softw. Technol., vol. 125, Sep. 2020, doi: 10.1016/j.infsof.2020.106330.

[8] P. V. Terlapu, K. K. Raju, G. Kiran Kumar, G. Jagadeeswara Rao, K. Kavitha, and S. Samreen, “Improved Software Effort Estimation Through Machine Learning: Challenges, Applications, and Feature Importance Analysis,” IEEE Access, vol. 12, pp. 138663–138701, 2024, doi: 10.1109/ACCESS.2024.3457771.

[9] M. Hosni, A. Idri, and A. Abran, “On the value of filter feature selection techniques in homogeneous ensembles effort estimation,” Journal of Software: Evolution and Process, vol. 33, no. 6, Jun. 2021, doi: 10.1002/smr.2343.

[10] H. D. P. De Carvalho, R. Fagundes, and W. Santos, “Extreme Learning Machine Applied to Software Development Effort Estimation,” IEEE Access, vol. 9, pp. 92676–92687, 2021, doi: 10.1109/ACCESS.2021.3091313.

[11] S. M. Mirjalili, S. M. Mirjalili, A. Lewis, M. Seyedali, M. Seyed Mohammad, and L. Andrew, “Grey Wolf Optimizer,” Advances in Engineering Software, vol. 69, pp. 46–61, 2014, doi: https://doi.org/10.1016/j.advengsoft.2013.12.007 Copyright.

[12] A. Idri, I. Abnane, and A. Abran, “Support vector regression-based imputation in analogy-based software development effort estimation,” Journal of Software: Evolution and Process, vol. 30, no. 12, Dec. 2018, doi: 10.1002/smr.2114.

[13] J. Kennedy, R. Eberhart, and bls gov, “Particle Swarm Optimization.”

[14] A. Puspaningrum, M. Mustamiin, F. Herdiyanti, and K. Noviyanto, “Software Effort Coefficient Optimization Using Hybrid Bat Algorithm and Whale Optimization Algorithm,” JURNAL INFOTEL, vol. 17, no. 1, pp. 122–135, May 2025, doi: 10.20895/infotel.v17i1.1250.

[15] A. G. Priya Varshini, K. Anitha Kumari, and V. Varadarajan, “Estimating software development efforts using a random forest-based stacked ensemble approach,” Electronics (Switzerland), vol. 10, no. 10, May 2021, doi: 10.3390/electronics10101195.

[16] A. Tripathi, K. K. Bharti, and M. Ghosh, “A fusion of binary grey wolf optimization algorithm with opposition and weighted positioning for feature selection,” International Journal of Information Technology (Singapore), vol. 15, no. 8, pp. 4469–4479, Dec. 2023, doi: 10.1007/s41870-023-01481-7.

[17] S. Chakraborty, A. K. Saha, R. Chakraborty, and M. Saha, “An enhanced whale optimization algorithm for large scale optimization problems,” Knowl. Based. Syst., vol. 233, Dec. 2021, doi: 10.1016/j.knosys.2021.107543.

[18] S. S. Gautam and V. Singh, “The state-of-the-art in software development effort estimation,” Journal of Software: Evolution and Process, vol. 30, no. 12, Dec. 2018, doi: 10.1002/smr.1983.

[19] H. Alsghaier and M. Akour, “Software fault prediction using Whale algorithm with genetics algorithm,” Softw. Pract. Exp., vol. 51, no. 5, pp. 1121–1146, May 2021, doi: 10.1002/spe.2941.

[20] D. Novitasari, I. Cholissodin, and W. F. Mahmudy, “Hybridizing PSO with SA for optimizing SVR applied to software effort estimation,” Telkomnika (Telecommunication Computing Electronics and Control), vol. 14, no. 1, pp. 245–253, Mar. 2016, doi: 10.12928/TELKOMNIKA.v14i1.2812.

[21] R. Rizkiana Putri and D. Candra Novitasari, “Hybrid GWO-PSO for Accurate Software Effort Estimation in COCOMO II,” Journal of Artificial Intelligence and Software Engineering, vol. 5, no. 3, pp. 1186–1192, 2025, doi: 10.30811/jaise.v5i3.7603.

[22] A. L. I. Oliveira, P. L. Braga, R. M. F. Lima, and M. L. Cornélio, “GA-based method for feature selection and parameters optimization for machine learning regression applied to software effort estimation,” Inf. Softw. Technol., vol. 52, no. 11, pp. 1155–1166, 2010, doi: 10.1016/j.infsof.2010.05.009.

[23] D. Rankovic, N. Rankovic, M. Ivanovic, and L. Lazic, “Convergence rate of Artificial Neural Networks for estimation in software development projects,” Inf. Softw. Technol., vol. 138, Oct. 2021, doi: 10.1016/j.infsof.2021.106627.