Comparative Analysis of Regression Techniques for Software Effort Estimation

• Published in: 8th International Conference on Recent Advances and Innovations in Engineering: Empowering Computing, Analytics, and Engineering Through Digital Innovation, ICRAIE 2023
• Main Author: Jayadi P.; Ahmad K.A.B.
• Format: Conference paper
• Language: English
• Published: Institute of Electrical and Electronics Engineers Inc. 2023
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85189935981&doi=10.1109%2fICRAIE59459.2023.10468186&partnerID=40&md5=fab6292617506967e322b07f6c82502f

Software effort estimation (SEE) is an essential aspect of software project management, influencing resource allocation and development schedules. While there are many approaches to SEE, choosing the suitable model remains a challenge. This study compared five regression models - Linear Regression, Ridge Regression, Lasso Regression, Decision Tree Regression, and Support Vector Regression - based on performance in estimating software effort using datasets COCOMO81 the evaluation results show that Lasso Regression has the lowest Mean Absolute Error (MAE) of 0.16902, while Decision Tree Regression recorded the highest MAE of 0.169784. Overall, the Lasso Regression stands out in its performance, with an R-squared of 51.76%, demonstrating its superior capabilities in feature selection. This research provides important insights into the effectiveness of regression models in the context of SEE, assisting practitioners in selecting appropriate methods for software projects. © 2023 IEEE.