Yazılım Maliyet Tahmininde Makine Öğrenimi ve Geleneksel Yöntemlerin Karşılaştırmalı Analizi

Year 2025, Volume: 8 Issue: 2, 120 - 140

Mehtap Ayal Serdar Solak

https://doi.org/10.51764/smutgd.1768788

Abstract

Yazılım projelerinde maliyet ve çaba tahmini, proje planlaması ve kaynak yönetimi açısından kritik bir öneme sahiptir. Geleneksel yöntemler yıllardır bu alanda kullanılmakta olsa da son yıllarda makine öğrenimi (ML) tabanlı modeller, doğruluk ve uygulanabilirlik açısından önemli avantajları bulunmaktadır. Bu çalışmada, yazılım geliştirme çaba tahmini alanında kullanılan geleneksel ve makine öğrenimi yöntemleri karşılaştırmalı olarak ele alınmıştır. Geleneksel yaklaşımların temel prensipleri açıklanmış, ardından makine öğrenimi tekniklerinin potansiyeli ve sınırlılıkları değerlendirilmiştir. Ayrıca, her iki yöntemin farklı veri setleri üzerindeki performansları incelenmiş, doğruluk, esneklik ve veri bağımlılığı gibi kriterler açısından karşılaştırmaları yapılmıştır. Sonuç olarak, tek bir yöntemin her veri setinde en iyi sonucu vermediği, ancak uygun veri ön işleme, özellik seçimi ve model yapılandırması ile makine öğrenimi tabanlı yaklaşımların geleneksel yöntemlere kıyasla daha üstün performans sergileyebildiği görülmüştür. Çalışma, yazılım çaba tahmini için model seçiminin proje bağlamına göre dikkatle yapılması gerektiğini vurgulamakta ve gelecekteki araştırmalar için çeşitli öneriler sunmaktadır.

Keywords

COCOMO , Geleneksel Yöntemler , Yazılım Maliyet Tahmini , Makine Öğrenimi Yöntemleri

Comparative Analysis of Machine Learning and Traditional Methods in Software Cost Estimation

Abstract

Cost and effort estimation in software projects is critically important for project planning and resource management. Although traditional methods have been applied in this field for many years, machine learning (ML)-based models have recently offered significant advantages in terms of accuracy and applicability. This study provides a comparative analysis of traditional and ML-based approaches in software effort estimation. The fundamental principles of traditional approaches are explained, followed by an evaluation of the potential and limitations of ML techniques. In addition, the performance of both approaches is examined on different datasets, comparing them in terms of accuracy, flexibility, and data dependency. The results indicate that no single method consistently produces the best outcome across all datasets; however, with proper data preprocessing, feature selection, and model configuration, ML-based approaches can outperform traditional ones. The study emphasizes that model selection for software effort estimation should be carried out carefully according to the project context and presents several recommendations for future research.

Keywords

COCOMO , Traditional Methods , Machine Learning Methods , Software Cost Estimation

References

• Alatawi, M. N. (2024). Forecasting the software engineering model’s effort estimation using constructive cost estimation models. Iran Journal of Computer Science, 7, 735–754. https://doi.org/10.1007/s42044- 024-00194-9
• Azzeh, M., Elsheikh, Y., & Alseid, M. (2014). An optimized analogy-based project effort estimation. International Journal of Computer Applications, 5(4). https://doi.org/10.48550/arXiv.1703.04563
• Biau, G., & Scornet, E. (2015). A random forest guided tour. arXiv preprint arXiv:1511.05741. https://arxiv.org/abs/1511.05741
• Boateng, E. Y., Otoo, J., & Abaye, D. A. (2020). Basic tenets of classification algorithms K-nearest-neighbor, support vector machine, random forest and neural network: A review. Journal of Data Analysis and Information Processing, 8, 341–357. https://www.scirp.org/journal/jdaip Boehm, B. W. (1981). Software engineering economics. Prentice-Hall.
• Boujida, F. E., Amazal, F. A., & Idri, A. (2025). Neural networks-based software development effort estimation: A systematic literature review. Journal of Software: Evolution and Process, 37, e2756. https://doi.org/10.1002/smr.2756
• Cervantes, J., Garcia-Lamont, F., Rodríguez-Mazahua, L., & Lopez, A. (2019). A comprehensive survey on support vector machine classification: Applications, challenges and trends. Neurocomputing. https://doi.org/10.1016/j.neucom.2018.06.079
• Chirra, S. M. R., & Reza, H. (2019). A survey on software cost estimation techniques. Journal of Software Engineering and Applications, 12, 226–248. https://doi.org/10.4236/jsea.2019.126014
• Cubillos, J., Aponte, J., Gomez, D., & Rojas, E. (2024). Agile effort estimation in Colombia: An assessment and opportunities for improvement. Science of Computer Programming, 236, 103115. https://doi.org/10.1016/j.scico.2024.103115
• Dashti, M., Gandomani, T. J., Adeh, D. H., Zulzalil, H., & Sultan, A. B. M. (2022). LEMABE: A novel framework to improve analogy-based software cost estimation using learnable evolution model. PeerJ Computer Science, 8, e800. http://dx.doi.org/10.7717/peerj-cs.800#supplemental-information
• Derya, D., Derya, O. B., & Dökeroğlu, T. (2024). Multi-objective software project cost estimation using recent machine learning approaches. Researcher, 4(1), 1–14.
• Draz, M. M., Emam, O., & Azzam, S. M. (2024). Software cost estimation prediction using a convolutional neural network and particle swarm optimization algorithm. Scientific Reports, 14, 13129. https://doi.org/10.1038/s41598-024-13129
• Eberendu, A. C. (2014). Software project cost estimation: Issues, problems and possible solutions. International Journal of Engineering Science Invention, 3(6), 38–43.
• Gandomani, T. J., Dashti, M., Ansaripour, S., & Zulzalil, H. (2025). Enhancing analogy-based software cost estimation using Grey Wolf Optimization algorithm. PeerJ Computer Science, 11, e2794. http://dx.doi.org/10.7717/peerj-cs.2794#supplemental-information
• Guevara-Vega, C. P., Basantes-Andrade, A., Guerrero-Pasquel, J., & Quiña-Mera, A. (2019). Software estimation: Benchmarking between COCOMO II and SCOPE. In Proceedings of the 4th International Conference on Information Technology Trends (CITT 2018) (pp. 167–179). Springer.
• Gunti, N. (2006). Function point analysis examined. Art of Software Engineering. Avenue A | Razorfish™. Gültekin, M., & Kalıpsız, O. (2016). Yapay sinir ağları tabanlı yazılım efor tahmini. Yönetim Bilişim Sistemleri Dergisi, 1(3), 246–253.
• Hammad, M., Alqaddoumi, A., Al-Obaidy, H., & Almseidein, K. (2019). Predicting software faults based on K- nearest neighbors classification. International Journal of Computing and Digital Systems, 8(5). http://dx.doi.org/10.12785/ijcds/080503
• Ismaeel, H. R., & Jamil, A. S. (2007). Software engineering cost estimation using COCOMO II model. Majallat al- Mansour, 10(86).
• Jadhav, A., & Shandilya, S. K. (2024). Towards effective feature selection in estimating software effort using machine learning. Journal of Software: Evolution and Process, 36, e2588. https://doi.org/10.1002/smr.2588
• Jadhav, A., Kaur, M., & Akter, F. (2022). Evolution of software development effort and cost estimation techniques: Five decades study using automated text mining approach. Mathematical Problems in Engineering, 2022, 5782587. https://doi.org/10.1155/2022/5782587
• Jadhav, A., Shandilya, S. K., Izonin, I., & Muzyka, R. (2024). Multi-step dynamic ensemble selection to estimate software effort. Applied Artificial Intelligence, 38(1), e2351718. https://doi.org/10.1080/08839514.2024.2351718
• Kara, Ş. E., & Şamlı, R. (2021). Yazılım projelerinin maliyet tahmini için WEKA’da makine öğrenmesi algoritmalarının karşılaştırmalı analizi. Avrupa Bilim ve Teknoloji Dergisi (European Journal of Science and Technology), 23, 415–426.
• Kayakuş, M. (2021). Yazılım çaba tahmininde yapay sinir ağları için optimum yapının belirlenmesi. European Journal of Science and Technology, Special Issue 22, 43–48.
• Li, Z., O'Brien, L. M., & Zhang, H. (2011). Circumstantial-evidence-based judgment for software effort estimation. In Proceedings of the International Conference on Software Engineering Research & Practice (SERP 2011) (pp. 18–27).
• Mahmood, Y., Kama, N., Azmi, A., Khan, A. S., & Ali, M. (2022). Software effort estimation accuracy prediction of machine learning techniques: A systematic performance evaluation. Software: Practice and Experience, 52(12), 2683–2706.
• Mansoor, F., Alim, M. A., Jilani, M. T., Alam, M. M., & Su’ud, M. M. (2024). Enhancing software cost estimation using feature selection and machine learning techniques. Computers, Materials & Continua, 79(3). https://doi.org/10.32604/cmc.2024.057979
• Matsubara, P. G. F., Steinmacher, I., Gadelha, B., & Conte, T. (2023). Much more than a prediction: Expert-based software effort estimation as a behavioral act. Empirical Software Engineering, 28, 98. https://doi.org/10.1007/s10664-023-10332-9
• Maulud, D. H., & Abdulazeez, A. M. (2020). A review on linear regression comprehensive in machine learning. Journal of Applied Science and Technology Trends, 1(2), 140–147.
• Meli, R., & Santillo, L. (1999). Function point estimation methods: A comparative overview. CiteSeer. https://www.researchgate.net/publication/2462027
• Mustafa, E. I., & Osman, R. (2024). A random forest model for early-stage software effort estimation for the SEERA dataset. Information and Software Technology, 169, 107413. https://doi.org/10.1016/j.infsof.2024.107413
• Mutale, B., Withanage, N. C., Mishra, P. K., Shen, J., Abdelrahman, K., & Fnais, M. S. (2024). A performance evaluation of random forest, artificial neural network, and support vector machine learning algorithms to predict spatio-temporal land use-land cover dynamics: A case from Lusaka and Colombo. Frontiers in Environmental Science, 12, 1431645. https://doi.org/10.3389/fenvs.2024.1431645
• Nassif, A. B., Azzeh, M., Capretz, L. F., & Ho, D. (2016). Neural network models for software development effort estimation: A comparative study. Applied Soft Computing, 62, 529–545.
• Nhung, H. L. T. K., Hai, V. V., Silhavy, R., Prokopova, Z., & Silhavy, P. (in press). Parametric software effort estimation based on optimizing correction factors and multiple linear regression. IEEE Access. Pisner, D. A., & Schnyer, D. M. (2022). Support vector machine. In Machine Learning (1st ed., pp. 101– 121). Elsevier Inc.
• Poženel, M., Fürst, L., Vavpotič, D., & Hovelja, T. (2023). Agile effort estimation: Comparing the accuracy and efficiency of planning poker, bucket system, and affinity estimation methods. International Journal of Software Engineering and Knowledge Engineering. https://doi.org/10.1142/S021819402350064X
• Qu, K. (2024). Research on linear regression algorithm. MATEC Web of Conferences, 395, 01046. https://doi.org/10.1051/matecconf/202039501046
• Roustaei, N. (2024). Application and interpretation of linear-regression analysis. Medical Hypothesis, Discovery & Innovation in Ophthalmology, 13(3), 151–159.
• Sánchez-García, Á. J., González-Hernández, M. S., Cortés-Verdín, K., & Pérez-Arriaga, J. C. (2024). Software estimation in the design stage with statistical models and machine learning: An empirical study. Mathematics, 12(7), 1058.
• Sardar, F., Latif, M. A., Khan, M. K., Al-Boridi, O., & Hamouda, H. (2025). Comparative analysis of AI models for effort estimation in Western and regional environments. International Journal of Mathematics, Statistics, and Computer Science, 3. https://doi.org/10.59543/ijmscs.v3i.15097
• Shadad, M. J., & Bahadar, F. (2017). Analyzing cost estimation model to optimize COCOMO II for enterprise level software. International Journal of Computer (IJC), 25(1), 168–179.
• Shailee, N. M., Alam, A., Ahmed, T., Rudro, R. A. M., & Nur, K. (2024). Software bug prediction using machine learning on JM1 dataset. In Proceedings of the International Conference on Advances in Computing, Communication, Electrical, and Smart Systems (iCACCESS 2024). IEEE.
• Shamim, M. M. I., Hamid, A. B. A., Nyamasvisva, T. E., & Rafi, N. S. B. (2025). Advancement of artificial intelligence in cost estimation for project management success: A systematic review of machine learning, deep learning, regression, and hybrid models. Modelling, 6(1), 35.
• Shdefat, A. Y., Mostafa, N., Al-Arnaout, Z., Kotb, Y., & Alabed, S. (2024). Optimizing HAR systems: Comparative analysis of enhanced SVM and k-NN classifiers. International Journal of Computational Intelligence Systems, 17, 150.
• Suyal, M., & Goyal, P. (2022). A review on analysis of K-nearest neighbor classification machine learning algorithms based on supervised learning. International Journal of Engineering Trends and Technology, 70(7), 43–48.
• Uc-Cetina, V. (2023). Recent advances in software effort estimation using machine learning. arXiv preprint arXiv:2303.03482. https://arxiv.org/abs/2303.03482
• Wen, J., Li, S., Lin, Z., Hu, Y., & Huang, C. (2012). Systematic literature review of machine learning based software development effort estimation models. Information and Software Technology, 54(1), 41–59.
• Yang, D., Wan, Y., Tang, Z., Wu, S., He, M., & Li, M. (2006). COCOMO-U: An extension of COCOMO II for cost estimation with uncertainty. In Q. Wang et al. (Eds.), Proceedings of SPW/ProSim 2006, Lecture Notes in Computer Science (Vol. 3966, pp. 132–141). Springer.
• Zagajewski, B., Kluczek, M., Raczko, E., Njegovec, A., Dabija, A., & Kycko, M. (2021). Comparison of random forest, support vector machines, and neural networks for post-disaster forest species mapping of the Krkonoše/Karkonosze transboundary biosphere reserve. Remote Sensing, 13(13), 2581.
• Zakrani, A., Hain, M., & Abdelwahed, N. (2019). Investigating the use of random forest in software effort estimation. In Proceedings of the Second International Conference on Intelligent Computing in Data Sciences (ICDS 2018) (pp. 343–352). Elsevier.
• Shamim, M. M. I., Hamid, A. B. b. A., Nyamasvisva, T. E., & Rafi, N. S. B. (2025). Advancement of artificial intelligence in cost estimation for project management success: A systematic review of machine learning, deep learning, regression, and hybrid models. Modelling, 6(2), 35. https://doi.org/10.3390/modelling6020035