Effort Estimation in Software Development Projects Using Supervised Machine Learning Techniques

A preliminary study was conducted to validate the use of machine learning for software effort estimation. Applying the CRISP-DM methodology to the ISBSG dataset, our findings show the Random Forest model was most effective. The analysis also identified key predictive factors, including software size, team productivity, and development platform. This work establishes a strong baseline for the current project, which will enhance these results by exploring advanced ensemble methods and newer datasets