OPTIMIZING SOFTWARE EFFORT ESTIMATION: A SYNERGISTIC HYBRID DEEP LEARNING FRAMEWORK WITH ENHANCED METAHEURISTIC OPTIMIZATION
DOI:
https://doi.org/10.55640/Keywords:
Software Effort Estimation (SEE), Deep Learning, Metaheuristic Optimization, rey Wolf Optimizer (GWO), G Hybrid Model, Convolutional Neural Network (CNN)Abstract
Background: Accurate Software Effort Estimation (SEE) remains one of the most critical and persistent challenges in software engineering. Traditional algorithmic models lack adaptability, while standard machine learning (ML) approaches require extensive feature engineering and often fail to capture complex, non-linear project dynamics. Deep Learning (DL) offers a promising alternative but is highly sensitive to hyperparameter configuration and architectural choices, making optimization a significant barrier.
Objective: This research proposes and validates a novel, synergistic hybrid framework, "Deep-MetaSEE," designed to overcome these limitations. The framework integrates a hybrid deep learning model (Conv-LSTM) with an Enhanced Grey Wolf Optimizer (EGWO).
Methods: The Conv-LSTM core is designed to automatically extract hierarchical spatial and temporal features from project data. The EGWO, an improved metaheuristic algorithm incorporating dynamic weighting and Lévy flight mechanisms, performs a multi-objective optimization to simultaneously identify the optimal feature subset, Conv-LSTM architecture (layers, nodes), and training hyperparameters (e.g., learning rate, dropout). The Deep-MetaSEE framework was rigorously evaluated on three public benchmark datasets (Desharnais, COCOMO81, Maxwell) and compared against traditional ML (Random Forest, SVR), standalone DL (ANN, CNN, LSTM), and other hybrid models (PSO-ANN, GWO-ANN) using standard metrics (MMRE, PRED(25), MAE).
Results: Empirical results demonstrate that the Deep-MetaSEE framework achieves statistically significant (p < 0.05) and superior estimation accuracy, consistently outperforming all baseline models across all datasets. The EGWO component also demonstrated faster convergence to a better optimal solution compared to standard GWO.
Conclusion: The proposed synergistic hybrid framework provides a robust and highly accurate solution for SEE. By automating feature selection and optimizing deep learning architecture simultaneously, Deep-MetaSEE addresses key gaps in current estimation literature and offers a powerful data-driven tool for project managers.
References
Arpteg A, Brinne B, Crnkovic-Friis L, Bosch J. 2018, August. Software engineering challenges of deep learning. In 2018 44th euromicro conference on software engineering and advanced applications (SEAA) (pp. 50–59). IEEE.
Azzeh M, Nassif AB, Martín CL. Empirical analysis on productivity prediction and locality for use case points method. Software Qual J. 2021;29:309–36.
Calleja A, Tapiador J, Caballero J. The malsource dataset: quantifying complexity and code reuse in malware development. IEEE Trans Inf Forensics Secur. 2018;14(12):3175–90.
Choetkiertikul M, Dam HK, Tran T, Pham T, Ghose A, Menzies T. A deep learning model for estimating story points. IEEE Trans Software Eng. 2018;45(7):637–56.
Curcio K, Navarro T, Malucelli A, Reinehr S. Requirements engineering: A systematic mapping study in agile software development. J Syst Softw. 2018;139:32–50.
Dey N, Hassanien E, Bhatt A, Ashour CS, A. and, Satapathy C. S., 2018. Internet of things and big data analytics toward next-generation intelligence. Springer Nature.
Gao W, Alsarraf J, Moayedi H, Shahsavar A, Nguyen H. Comprehensive preference learning and feature validity for designing energy-efficient residential buildings using machine learning paradigms. Appl Soft Comput. 2019;84:105748.
García-Martín E, Rodrigues CF, Riley G, Grahn H. Estimation of energy consumption in machine learning. J Parallel Distrib Comput. 2019;134:75–88.
Iglhaut J, Cabo C, Puliti S, Piermattei L, O’Connor J, Rosette J. Structure from motion photogrammetry in forestry: A review. Curr Forestry Rep. 2019;5:155–68.
Jain NK, Celo S, Kumar V. Internationalization speed, resources and performance: evidence from Indian software industry. J Bus Res. 2019;95:26–37.
Jayanthi R, Florence L. Software defect prediction techniques using metrics based on neural network classifier. Cluster Comput. 2019;22:77–88.
Kaushik A, Singal N. 2019. A hybrid model of wavelet neural network and metaheuristic algorithm for software development effort Estimation. Int J Inform Technol, 1–10.
Khan MS, ul Hassan CA, Shah MA, Shamim A. 2018, September. Software cost and effort estimation using a new optimization algorithm inspired by strawberry plant. In 2018 24th International Conference on Automation and Computing (ICAC) (pp. 1–6). IEEE.
Kumar PS, Behera HS, Kumari A, Nayak J, Naik B. Advancement from neural networks to deep learning in software effort estimation: perspective of two decades. Comput Sci Rev. 2020;38:p100288.
Liu C, Yang D, Xia X, Yan M, Zhang X. A two-phase transfer learning model for cross-project defect prediction. Inf Softw Technol. 2019;107:125–36.
Mohanani R, Salman I, Turhan B, Rodríguez P, Ralph P. Cognitive biases in software engineering: a systematic mapping study. IEEE Trans Software Eng. 2018;46(12):1318–39.
Moosavi SHS, Bardsiri VK. Poor and rich optimization algorithm: A new human-based and multi populations algorithm. Eng Appl Artif Intell. 2019;86:165–81.
Mustapha H, Abdelwahed N. Investigating the use of random forest in software effort Estimation. Procedia Comput Sci. 2019;148:343–52.
Pospieszny P, Czarnacka-Chrobot B, Kobylinski A. An effective approach for software project effort and duration Estimation with machine learning algorithms. J Syst Softw. 2018;137:184–96.
Qamar N, Batool F, Zafar K. Efficient effort Estimation of web based projects using neuro-web. Int J Adv Appl Sci. 2018;5(11):33–9.
Rafique W, Qi L, Yaqoob I, Imran M, Rasool RU, Dou W. Complementing IoT services through software defined networking and edge computing: A comprehensive survey. IEEE Commun Surv Tutorials. 2020;22(3):1761–804.
Rathore SS, Kumar S. A study on software fault prediction techniques. Artif Intell Rev. 2019;51:255–327.
Schön, E. M., Thomaschewski, J., & Escalona, M. J. (2017). Agile Requirements Engineering: A systematic literature review. Computer standards & interfaces, 49, 79-91.
Silhavy R, Silhavy P, Prokopova Z. Evaluating subset selection methods for use case points Estimation. Inf Softw Technol. 2018;97:1–9.
Singh, V. (2024). The impact of artificial intelligence on compliance and regulatory reporting. J. Electrical Systems, 20(11s), 4322–4328. https://doi.org/10.52783/jes.8484
Tam C, da Costa Moura EJ, Oliveira T, Varajão J. The factors influencing the success of on-going agile software development projects. Int J Project Manage. 2020;38(3):165–76.
Zakrani A, Idri A, Hain M. 2020. Software effort estimation using an optimal trees ensemble: An empirical comparative study. In Proceedings of the 8th International Conference on Sciences of Electronics, Technologies of Information and Telecommunications (SETIT’18), Vol. 1 (pp. 72–82). Springer International Publishing.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 Linh Thuy Nguyen, Kofi Mensah (Author)
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors retain the copyright of their manuscripts, and all Open Access articles are disseminated under the terms of the Creative Commons Attribution License 4.0 (CC-BY), which licenses unrestricted use, distribution, and reproduction in any medium, provided that the original work is appropriately cited. The use of general descriptive names, trade names, trademarks, and so forth in this publication, even if not specifically identified, does not imply that these names are not protected by the relevant laws and regulations.